Aligning the decision-making process of deep learning models with that of experienced sonographers is essential for ultrasound-based reliable disease diagnosis. Although existing methods have made significant progress in this aspect, their alignments are primarily associational rather than causal, leading to pseudo-correlations between features and diagnostic results. Such a biased diagnosis blindly models the sonographer's diagnostic skills and attention to specific patterns, which we argue hardly produces an AI diagnoser that is comparable to human experts. To address this issue, we propose a causality-based diagnostic framework to align the model's diagnostic behaviors with those of experts. Specifically, by delving into both conspicuous and inconspicuous confounders within the ultrasound images, the back-door and front-door adjustment causal learning modules are proposed to promote unbiased learning by mitigating potential pseudo-correlations. In addition, we integrate causal inference into a well-designed dual-branch model with feature interaction bridges for compatibility with multimodal ultrasound inputs. To fully evaluate our method, we conduct comparative studies on different diseases and ultrasound modalities. In particular, we publish a carefully constructed multimodal ultrasound dataset for breast lesion diagnosis and segmentation. Sufficient comparative and ablation studies on this dataset emphasize that our method outperforms state-of-the-art methods.

Since a building's floorplan remains consistent over time and is inherently robust to changes in visual appearance, visual Floorplan Localization (FLoc) has received increasing attention from researchers. However, as a compact and minimalist representation of the building's layout, floorplans contain many repetitive structures (e.g., hallways and corners), thus easily result in ambiguous localization. Existing methods either pin their hopes on matching 2D structural cues in floorplans or rely on 3D geometry-constrained visual pre-trainings, ignoring the richer contextual information provided by visual images. In this paper, we suggest using broader visual scene context to empower FLoc algorithms with scene layout priors to eliminate localization uncertainty. In particular, we propose an unsupervised learning technique with clustering constraints to pre-train a room discriminator on self-collected unlabeled room images. Such a discriminator can empirically extract the hidden room type of the observed image and distinguish it from other room types. By injecting the scene context information summarized by the discriminator into an FLoc algorithm, the room style knowledge is effectively exploited to guide definite visual FLoc. We conducted sufficient comparative studies on two standard visual Floc benchmarks. Our experiments show that our approach outperforms state-of-the-art methods and achieves significant improvements in robustness and accuracy.

Accurately detecting super host that establishes connections to a large number of distinct peers is significant for mitigating web attacks and ensuring high quality of web service. Existing sketch-based approaches estimate the number of distinct connections called flow cardinality according to full IP addresses, while ignoring the fact that a malicious or victim super host often communicates with hosts within the same subnet, resulting in high false positive rates and low accuracy. Though hierarchical-structure based approaches could capture flow cardinality in subnet, they inherently suffer from high memory usage. To address these limitations, we propose SegSketch, a segmented cardinality estimation approach that employs a lightweight halved-segment hashing strategy to infer common prefix lengths of IP addresses, and estimates cardinality within subnet to enhance detection accuracy under constrained memory size. Experiments driven by real-world traces demonstrate that, SegSketch improves F1-Score by up to 8.04× compared to state-of-the-art solutions, particularly under small memory budgets.

MOTIVATION: Identifying synergistic drug combinations holds promise for more effective treatment strategies. Recent deep learning methods such as Transformers and Graph Neural Networks have shown improved predictive performance, but most of them integrate drug and cell line representations without explicitly modelling the causal effects of genes in mediating drug responses. RESULTS: We introduce CADS (Causal Adjustment for Drug Synergy), a deep learning framework that explicitly models the gene-drug causal relationships to improve both prediction accuracy and biological interpretability. CADS integrates multi-omics data with a learnable gene-selection mechanism that performs causal backdoor adjustment, enabling both drug synergy prediction and causal gene discovery. Across multiple benchmark datasets, CADS consistently achieves superior performance compared with state-of-the-art drug synergy prediction models. In addition, downstream analyses on case studies demonstrate that the inferred gene causal scores can recover clinically validated cancer-related genes involved in drug combinations. These results demonstrate that explicitly modelling causal genetic effects can enhance the reliability and interpretability of drug synergy prediction. AVAILABILITY AND IMPLEMENTATION: The source code of CADS can be found at https://github.com/HuaiwuZhang/causalDC.

By generating massive gene transcriptome data and analyzing transcriptomic variations at the cell level, single-cell RNA-sequencing (scRNA-seq) technology has provided new way to explore cellular heterogeneity and functionality. Clustering scRNA-seq data could discover the hidden diversity and complexity of cell populations, which can aid to the identification of the disease mechanisms and biomarkers. In this paper, a novel method (DSINMF) is presented for clustering single cell RNA sequencing data by using deep matrix factorization. Our proposed method comprises four steps: first, the feature selection is utilized to remove irrelevant features. Then, the dropout imputation is used to handle missing value problem. Further, the dimension reduction is employed to preserve data characteristics and reduce noise effects. Finally, the deep matrix factorization with bi-stochastic graph regularization is used to obtain cluster results from scRNA-seq data. We compare DSINMF with other state-of-the-art algorithms on nine datasets and the results show our method outperformances than other methods. The code can be downloaded from https://github.com/lanbiolab/DSINMF.

The arduous and costly journey of drug discovery is increasingly intersecting with computational approaches, which promise to accelerate the analysis of bioassays and biomedical literature. The critical role of microRNAs (miRNAs) in disease progression has been underscored in recent studies, elevating them as potential therapeutic targets. This emphasizes the need for the development of sophisticated computational models that can effectively identify promising drug targets such as miRNAs. Herein, we present a novel method, termed Duplex Link Prediction (DLP), rooted in subspace segmentation, to pinpoint potential miRNA targets. Our approach initiates with the application of the Network Enhancement (NE) algorithm to refine the similarity metric between miRNAs. Thereafter, we construct two matrices by pre-loading the association matrix from both the drug and miRNA perspectives, employing the K Nearest Neighbors (KNN) technique. The DLSR algorithm is then applied to predict potential associations. The final predicted association scores are ascertained through the weighted mean of the two matrices. Our empirical findings suggest that the DLP algorithm outperforms current methodologies in the realm of identifying potential miRNA drug targets. Case study validations further reinforce the real-world applicability and effectiveness of our proposed method.

Brain disorders are associated with impairments in cognitive and social functioning, placing a substantial burden on families, healthcare systems, and communities. However, accurate diagnosis remains challenging due to complex higher order interactions among brain regions. Existing graph-based methods are largely limited to pairwise connectivity. In addition, these methods often fail to fully exploit well-established neuroscientific prior knowledge, resulting in limited biological interpretability and suboptimal diagnostic performance. Therefore, we propose a prior knowledge-guided adaptive hypergraph learning (PK-AHGL) framework that represents individual-level functional connectivity networks as hypergraphs to capture higher order multiregion interactions while incorporating neuroscientific prior knowledge for brain disorder diagnosis. PK-AHGL consists of three key modules: 1) an adaptive hypergraph convolution module. Unlike traditional hypergraph neural networks that use static hyperedge weights, this module adaptively learns the weights of different hyperedges; 2) a sparse affinity Laplacian module. Key brain regions are extracted from disorder related functional brain networks and used as prior knowledge. Based on these regions, we compute a hyperedge similarity matrix that encourages similar hyperedges to have similar weights; and 3) a proportional margin ranking module. This module further utilizes prior knowledge by guiding hyperedges containing a higher proportion of key brain regions to obtain larger weights. Experiments on autism brain imaging data exchange (ABIDE), Strategic Research Program for the Promotion of Brain Science (SRPBS)-schizophrenia (SCZ), SRPBS-major depressive disorder (MDD), and Alzheimer’s disease neuroimaging initiative (ADNI) show that PK-AHGL achieves accuracies of 75.78%, 82.66%, 74.89%, and 77.22%, respectively, outperforming multiple state-of-the-art methods. These results suggest that PK-AHGL provides an effective auxiliary tool for brain disorder diagnosis and may support community-oriented mental health services.

Survival risk prediction is crucial for personalized treatment of gliomas. Medical image foundational models can explore complex medical features, which are critical for prognosis in gliomas. We propose SAM-Risk, which uses a customized SAM-Med3D with multi-view representation fusion and clinical knowledge-based age-grade stratified loss for glioma survival risk prediction. First, to utilize potential interactions between multiple views at an early stage, we design a 3D representation generation module that transforms 1D handcrafted radiomics and clinical features into 3D representations, which are fused with multimodal MRIs through a multi-view representation fusion module. The fused representation is fed into the customized SAM-Med3D, fine-tuned using LoRA and a disparity function to extract survival risk-related features. We design a feature refinement module to explore the inter-channel relationships among the outputs of the fine-tuned SAM-Med3D. Additionally, we propose an age-grade stratified loss based on glioma prognosis standards to make the predicted risk more consistent with clinical prior knowledge. Validated on two publicly available UCSF-PDGM and BraTS2020 datasets, SAM-Risk achieves a C-index of 75.08% and 73.67%, respectively, outperforming several survival risk prediction methods.

Many deep learning methods represented by graph-based approaches achieve significant progress in drug repositioning. However, these graph-based methods face a critical limitation: they often fail in cold-start scenarios because the graph structure relies heavily on known association information from both the drug and disease sides. To address this challenge, we propose a bidirectional behavior learning strategy for drug repositioning, BiBLDR, an innovative framework that reformulates drug repositioning as a behavior sequence learning task. First, we construct bidirectional behavioral sequences based on drug and disease sides. Bidirectional behavior sequences ensure sufficient information for model learning in both drug and disease cold-start scenarios, while providing more precise feature representations for association prediction tasks. Subsequently, we propose a two-stage strategy for drug repositioning. In the first stage, we construct prototype spaces to characterise the representational attributes of drugs and diseases. In the second stage, these refined prototypes and bidirectional behavior sequence data are leveraged to predict potential drug-disease associations. This design allows BiBLDR to more robustly capture hidden pharmacological relationships from bidirectional behavioral sequences, delivering significant benefits in cold-start scenarios. Extensive experiments demonstrate that our method achieves state-of-the-art performance on benchmark datasets. Meanwhile, BiBLDR demonstrates significantly superior performance compared to previous methods in cold-start scenarios.

Accurately predicting drug-drug interaction events (DDIEs) is critical for improving medication safety and guiding clinical decision-making. However, existing graph neural network (GNN)-based methods often struggle to effectively integrate multi-view features and generalize to novel or understudied drugs. To address these limitations, we propose MRLF-DDI, a multi-view representation learning framework that jointly models information from individual drug features, local interaction contexts, and global interaction patterns. MRLF-DDI introduces the use of atom-level structural features enriched with bond angle information-marking the first incorporation of this geometry-aware feature in DDIE prediction. It further employs a multi-granularity GNN and a gated knowledge transfer strategy to enhance feature learning and cold-start generalization. Extensive experiments on benchmark datasets demonstrate that MRLF-DDI achieves superior performance in both warm-start and cold-start scenarios. Case studies and visualization analyses further highlight its practical utility in identifying clinically relevant DDIEs.

Object navigation (ObjcetNav), which enables an agent to seek any instance of an object category, has shown great advances. However, current agents are built upon occlusion-prone visual observations or compressed 2-D maps, which hinder their embodied perception of 3-D scene geometry. Furthermore, existing methods usually decouple ObjectNav into the exploration and exploitation subtasks, easily leading to ambiguous object localization and blind exploration. To address these issues, we first propose an embodied contrastive learning (ECL) method with geometric consistency (GC) and behavioral awareness (BA), which motivates agents to encode 3-D scene layouts and semantic cues actively. The BA is modeled by predicting navigational actions based on multiframe visual images, as behaviors causing differences between adjacent visual sensations are crucial for learning correlations among continuous visions. The GC is modeled by aligning the behavior-aware visual stimulus with 3-D semantic shapes through unsupervised contrastive learning. Then, based on the above ECL pretraining, a coarse-to-fine ObjectNav policy with explorer and discriminator cooperation is proposed, inspired by the treasure-hunting mindset. Concretely, the explorer is designed to adaptively switch the action spaces, thereby switching the global and local exploration thoughts according to the accumulated scene priors. The discriminator is designed to discriminate the target's authenticity using behavior-aware visual features and geometric invariance priors, which permits mimicking the human behavior of "approaching to confirm" when distinguishing objects from a distance. As expected, our ECL method performs well on object detection (ObjDet) and instance segmentation (InstSeg) tasks. Our ECL-enhanced ObjectNav strategy outperforms state-of-the-art (SOTA) methods on Matterport3D (MP3D), Gibson, and HM3D datasets.

Predicting drug-target interactions (DTIs) plays an essential role in drug discovery and drug repurposing. Although significant performance improvements have been achieved in DTI prediction, existing methods have not fully explored the properties of protein and drug molecular structure to make results interpretable. In this study, we propose MotifGT-DTI, a novel motif-based model with a graph transformer (GT) for DTI prediction. Specifically, MotifGT-DTI captures complex molecular patterns of drug molecular graph motifs and protein 3-D pocket subgraphs with GT. To attain protein characteristics more comprehensively, MotifGT-DTI fuses 1-D sequence and 3-D structure features with cross-attention from two views. Then, the structural-level association patterns of drug molecules and proteins are connected via a bilinear attention network. Experimental results show that MotifGT-DTI achieves the best accuracy compared to state-of-the-art baselines on four public datasets. In the three cold-start scenarios, the prediction results provided by our method are competitive in accuracy, generalization ability, and stability, highlighting its promising potential for practical applications. Furthermore, the visualization study demonstrates that MotifGT-DTI finds functional molecular motifs and provides interpretability for predicted results. The datasets and codes are publicly available at https://github.com/Dimpleney/MotifGT-DTI.

Remote Direct Memory Access (RDMA) over Converged Ethernet (RoCE) cooperating with Priority Flow Control (PFC) has been widely deployed in production datacenters to enable low latency, lossless transmission. At the same time, modern datacenters typically offer parallel transmission paths between any pair of end-hosts, underscoring the importance of load balancing. However, the well-studied load balancing mechanisms designed for lossy datacenter networks (DCNs) are ill-suited for such lossless environments. Through extensive experiments, we are among the first to comprehensively inspect the interactions between PFC and load balancing, and uncover that existing fine-grained rerouting schemes can be counterproductive to spread the congested flows among more paths, further aggravating PFC’s head-of-line (HoL) blocking. Motivated by this, we present FLB, a Fine-grained Load Balancing scheme for lossless DCNs. At its core, FLB employs threshold-free rerouting to effectively balance traffic load and improve link utilization during normal conditions and leverages timely congested flow isolation to eliminate HoL blocking on non-congested flows when congestion occurs. To handle complex multi-bottleneck scenarios, we further introduce FLB*, which incorporates an enhanced congestion-point-aware isolation mechanism using Congestion Point Identifiers (CPI) to eliminate HoL blocking among different congested flows.We have fully implemented a FLB prototype, and our evaluation results show that FLB reduces PFC PAUSE rate by up to 96% and avoids HoL blocking, translating to up to 45% improvement in goodput over CONGA+DCQCN and 40%, 36%, 29% and 18% reduction in average flow completion time (FCT) over LetFlow+Swift, MP-RDMA, Proteus+DCQCN and LetFlow+PCN, respectively.

In recent years, end-to-end congestion control algorithms or flow pausing mechanisms are proposed to achieve high throughput and low latency in datacenter networks. However, prior end-to-end congestion control works without complex signals fail to achieve fast convergence to a stable equilibrium state and effectively handle the transient congestion, while existing flow pausing mechanisms are decoupled from congestion control, which leads to long convergence time after transient states and incomplete queue elimination in equilibrium states. To address these issues, we present FAR, a rate control protocol that combines the advantages of flow pausing and congestion control. At its heart, FAR couples the bandwidth-estimation-based congestion control and the end-to-end flow pausing mechanisms. After flow pausing, FAR quickly explore the available bandwidth with a binary-search probe to achieve high throughput and low latency. Meanwhile, FAR employs a probe staggering mechanism to address the queue oscillation issue in high-concurrency scenarios. We implement the prototype of FAR using DPDK. Extensive evaluation results demonstrate that our protocol achieves accurate bandwidth estimation and reduces the tail flow completion time (FCT) by up to 67% compared with the state-of-the-art designs.

With the growing ubiquity of video streaming, ensuring a fair and high quality of experience (QoE) for users has emerged as a shared concern among video content providers. State-of-the-art video delivery systems achieve QoE fairness through bottleneck bandwidth allocation across multiple video streams, all based on the assumption of a unified congestion control (CC) protocol. However, the widespread use of heterogeneous CC protocols on the Internet not only disrupts QoE fairness among video streams but also poses challenges in achieving fast convergence under dynamic bandwidth. To address these issues, we propose a QoE-Fairnessawarebandwidthallocationmechanism called Fabam, which establishes a unified QoE control plane across heterogeneous CC protocols. Fabam constructs independent virtual targets based on the real-time QoE of each video stream to achieve QoE fairness, and offers rapid convergence for the underlying CC protocols to improve efficiency. In addition, we propose a Deep Neural Network (DNN)-based multi-step mapping model aimed at balancing the performance and overhead of Fabam, thereby enhancing its deployment potential in practical applications. We implement Fabam on QUIC and integrate it with Dash.js. The evaluation results demonstrate the significant superiority of Fabam over the state-of-the-art approaches, including an enhancement of 44.01% in QoE fairness and an improvement of 36.39% in QoE efficiency. Meanwhile, Fabam-DNN maintains satisfactory QoE fairness while supporting multiple users at a low cost.

Synchronous stochastic gradient descent (SSGD) has been widely used in distributed deep learning. However, since the local gradients need to be shared among workers at every iteration, SSGD performance is significantly influenced by network bottlenecks caused by either heterogeneous environment or bandwidth contention. To solve this problem, asynchronous parallel (ASP) strategy allows each worker to update parameters independently without synchronization, while suffering from accuracy loss and convergence inefficiency. In this paper, we propose a novel similarity-based synchronization scheme called SIM, which mitigates the impact of network bottlenecks and ensures convergence efficiency. Specifically, SIM reduces the number of aggregation workers based on the gradient similarity between global and local gradients, therefore shrinking the waiting time for the stragglers. We provide a theoretical analysis of convergence efficiency and conduct large-scale testbed experiments on CIFAR-10 and SQUAD dataset. The experimental results show that SIM reduces the convergence time of four classical deep learning models by up to 40%.

In this paper, we consider the k-center problem with outliers (the (k, z)-center problem) in the context of Massively Parallel Computation (MPC). Existing MPC algorithms for the (k, z)-center problem typically require Ω(k) local space per machine. While this may be feasible when k is small, these algorithms become impractical for large k, where each machine may lack sufficient space for computation. This motivates the study of fully-scalable algorithms with sublinear local space. We propose the first fully-scalable MPC algorithm for the (k, z)-center problem. The main challenge is to design an MPC algorithm that operates with sublinear local space for finding the inliers close to the optimal clustering centers, and ensuring the approximation loss remains bounded. To address this issue, we propose an iterative sampling-based algorithm with sublinear local space in the data size. A key component of our approach is an outliers-removal algorithm that adjusts the sample size in each iteration to select inliers as clustering centers. However, the number of discarded inliers increases with the iteration of the outliers-removal algorithm, making it difficult to bound. To address this, we propose a self-adaptive method that can automatically adjust sample size to account for different data distributions on each machine, ensuring a lower bound on the sampling success probability. With these techniques, we present an O(log^*n)-approximation MPC algorithm for the (k, z)-center problem in constant-dimensional Euclidean space. The algorithm discards at most (1 + ε)z outliers, completing in O(log log n) computation rounds while using Θ(n^δ) local space per machine.

Identifying cancer genes is crucial for treatment and understanding pathogenesis. Recent methods typically leverage protein-protein interaction (PPI) networks or gene functional association data from annotated gene sets. There may be some shared neighborhood structure information between these two types of gene association data. While this common information may contain more accurate gene association information, existing methods often overlook this potential. To address this gap, we introduce DISFusion, which integrates multi-omics cancer data, PPI networks, and gene functional associations to identify cancer genes. A key innovation of DISFusion is the cross-view decorrelation loss, which enhances the common information between PPI networks and gene functional associations, thereby improving prediction accuracy. Extensive experiments indicate that DISFusion outperforms state-of-the-art methods and exhibits greater generalization ability. Moreover, analysis of CPTAC pan-cancer proteomic data highlights significant associations between the 30 novel cancer genes predicted by DISFusion and multiple cancer types, underscoring its practical utility. These findings validate the effectiveness of enhancing common information and provide new insights into cancer gene identification.

Unified information extraction (UIE) aims to extract diverse structured information from unstructured text. While large language models (LLMs) have shown promise for UIE, they require significant computational resources and often struggle to generalize to unseen tasks. We propose RUIE (Retrieval-based Unified Information Extraction), a framework that leverages in-context learning for efficient task generalization. RUIE introduces a novel demonstration selection mechanism combining LLM preferences with a keyword-enhanced reward model, and employs a bi-encoder retriever trained through contrastive learning and knowledge distillation. As the first trainable retrieval framework for UIE, RUIE serves as a universal plugin for various LLMs. Experimental results on eight held-out datasets demonstrate RUIE’s effectiveness, with average F1-score improvements of 19.22 and 3.22 compared to instruction-tuning methods and other retrievers, respectively.

The congestion control (CC) algorithm is expected to achieve consistent high performance under different network environments. Traditionally, classic CCs are designed with the methodology of inferring path conditions to guide the rate adjustment. However, this methodology suffers from wrong path condition inferences in certain cases, which mislead the rate adjustment and lead to performance degradation. To avoid wrong path condition inferences, we develop the projection-based introspective method and design the introspective congestion control (ICC) algorithm in this paper. Specifically, the rate adjustment rules are designed to possess a specialized profile such that the projection of the profile can be distinguished under unchanged path conditions. In this way, the projection, which can be distinguished from the time series of delay signals in the frequency domain, facilitates ICC to extract more information for path condition inferences. Consequently, with the introspection on the projection, ICC can avoid being misled by wrong path condition inferences and thus achieve consistent high performance under different conditions. The advantages of ICC are confirmed through extensive experiments conducted on various locally emulated scenarios, global testbeds over the Internet, and the Alipay platform.

In this paper, we study the clustering problems in the learning-augmented setting, where predicted labels for a d-dimensional dataset with size m are given by an oracle to serve as auxiliary information to improve the clustering performance. Following the prior work, the given oracle is parameterized by some error rate α, which captures the accuracy of the oracle such that there are at most α fraction of false positives and false negatives in each predicted cluster. In this setting, the goal is to design fast and practical algorithms that can break the computational barriers of inapproximability. The current state-of-the-art learning-augmented k-means algorithm relies on sorting strategies to find good coordinates approximation, where a (1+O(α))-approximation can be achieved with near-linear running time in the data size. However, the computational demands for sorting may limit the scalability of the algorithm for handling large-scale datasets. To address this issue, in this paper, we propose new algorithms that can identify good coordinates approximation using sampling-based strategies, where (1+O(α))-approximation can be achieved with linear running time in the data size. To obtain a more practical algorithm for the problem with better clustering quality and running time, we propose a sampling-based heuristic which can directly find center approximations using sampling-based strategies. Empirical experiments show that our proposed methods are faster than the state-of-the-art learning-augmented k-means algorithms with comparable performances on clustering quality.

Fair clustering has recently received considerable attention where numerous distinct fairness notions are developed. Despite being well-justified, these fairness notions are frequently studied in isolation, leaving the need to explore how they can be combined. Building on prior work, we focus on the doubly constrained fair clustering that incorporates two widely adopted demographic representation fairness notions in clustering: group fairness and data summarization fairness. Both fairness notions extend classical clustering formulation by associating each data point with a demographic label, where group fairness requires each cluster to proportionally reflect the population-level distribution of demographic groups, and data summarization fairness ensures the chosen facilities maintaining the population-level demographic representation of each group. In this paper, we study the Fixed-Parameter Tractable (FPT) approximation algorithms for doubly constrained fair clustering under the k-median objective, referred to Df-k-Med. The previous algorithms typically enumerate different demographic groups or construct fairness coreset, parameterized by both the number of opened facilities and demographic labels. By further leveraging the local fairness information, we propose a color-agnostic structural method that obtains the parameterized result independent of the number of demographic labels while effectively handling the combination of both fairness constraints. Specifically, we design a constant factor approximation for the Df-k-Med problem with fairness violation by one, which runs in FPT(k)-time, where k is the number of opened facilities.

As the foundational components of large-scale applications, distributed key-value stores must respond to user requests quickly. However, a user request typically comprises multiple key-value access operations, which are processed in parallel across different servers, and the response time is determined by the slowest operation. To reduce the mean response time of requests, existing approaches schedule the sequence of key-value access operations across different servers so that all operations of a request complete at approximately the same time. Nevertheless, all of these approaches operate in a distributive manner, and their theoretical performance boundaries are unknown. To address these issues, we designed SDN-KVS (Key-Value Scheduler based on Software Defined Network), which migrates the waiting queue of key-value access operations from overloaded servers to the SDN controller. In this way, SDN-KVS centrally schedules the requests from different clients to overloaded servers without extra latency overhead. The scheduling result is proven to be$(1+2 \eta)$-approximation, i.e., the mean response time of requests is smaller than ($1+2 \eta$) times of the optimal value, where$\eta$is the parameter to make a trade-off between mean and tail response time. Simulation results confirm the excellent performance of the SDN-KVS algorithm. Specifically, SDN-KVS outperforms existing algorithms up to 37.6% and 79.8% in terms of mean and tail response time, respectively.

Distributed deep learning has been widely deployed in data centers to provide various services such as image classification and speech recognition. To reduce the training time, Top-k compression has become one of the most popular solutions used to shrink the data volume of gradients. Nevertheless, we observe that existing Top-k compression solutions are inefficient when used for large-scale distributed training due to gradient build-up, missing of Top-k gradients, and high compression overhead at the end hosts. To address these problems, we propose SwitchTop-k, which improves the accuracy of selecting Top-k values while ensuring a high compression rate and zero compression overhead. Specifically, SwitchTop-k offloads the Top-k compression from the end hosts to the programmable switches, thus alleviating the gradient build-up and compression overhead. Meanwhile, we propose a sketch-based solution to achieve high accuracy in selecting global Top-k gradients. We also co-design switch logic and end host logic to improve communication efficiency of uncompressed traffic. Finally, we implement SwitchTop-k on Intel Tofino switches and integrate it with Pytorch. The test results show that SwitchTop-k reduces iteration time by up to 91% compared with existing compression algorithms.

Since a building's floorplans are easily accessible, consistent over time, and inherently robust to changes in visual appearance, self-localization within the floorplan has attracted researchers' interest. However, since floorplans are minimalist representations of a building's structure, modal and geometric differences between visual perceptions and floorplans pose challenges to this task. While existing methods cleverly utilize 2D geometric features and pose filters to achieve promising performance, they fail to address the localization errors caused by frequent visual changes and view occlusions due to variously shaped 3D objects. To tackle these issues, this paper views the 2D Floorplan Localization (FLoc) problem from a higher dimension by injecting 3D geometric priors into the visual FLoc algorithm. For the 3D geometric prior modeling, we first model geometrically aware view invariance using multi-view constraints, i.e., leveraging imaging geometric principles to provide matching constraints between multiple images that see the same points. Then, we further model the view-scene aligned geometric priors, enhancing the cross-modal geometry-color correspondences by associating the scene's surface reconstruction with the RGB frames of the sequence. Both 3D priors are modeled through self-supervised contrastive learning, thus no additional geometric or semantic annotations are required. These 3D priors summarized in extensive realistic scenes bridge the modal gap while improving localization success without increasing the computational burden on the FLoc algorithm. Sufficient comparative studies demonstrate that our method significantly outperforms state-of-the-art methods and substantially boosts the FLoc accuracy.

Local search is a powerful clustering technique that provides high-quality solutions with theoretical guarantees. With distance-based sampling strategies, local search methods can achieve constant approximations for clustering with linear running time in data size. Despite their effectiveness, existing algorithms still face scalability issues as they require scanning the entire dataset for iterative center swaps. This typically leads to an O(ndk) running time, where n is the data size, d is the dimension, k is the number of clusters. To further improve the efficiency of local search algorithms, we propose new methods based on adaptive sampling and bandit strategies. Specifically, adaptive sampling can well approximate the distance-based sampling distribution without maintaining pairwise distances between data points and the centers, enabling fast and accurate sampling in sublinear time after an $\tilde{O}(nd)$ time preprocessing step. The bandit strategy models the best swap pair selection as a bandit problem, where a grouping strategy is proposed for fast identification of the optimal swap pair. With these techniques, our proposed algorithm can achieve constant approximation in expected running time $\tilde{O}(nd + k^4)$ under mild assumptions on optimal clusters and swap pair distributions. Our approach also extends naturally to the k-median objective, achieving constant approximation in expected running time $\tilde{O}(nd + \sqrt{n}k^3)$ without distributional assumptions. Empirical results demonstrate that our algorithm achieves up to 1000× speedup over existing local search methods on datasets with 100 million points, while delivering comparable clustering quality. Compared to coreset-based approaches, it provides up to around 80× speedup and consistently yields better clustering results.

Clustering is a fundamental problem that has been extensively studied over past few decades, with most research focusing on point-based clustering such as $k$-means, $k$-median, and $k$-center. However, numerous real-world applications, such as motion analysis, computer vision, and missing data analysis, require clustering over structured data, including lines, time series and affine subspaces (flats), where traditional point-based clustering algorithms often fall short. In this paper, we study the $k$-means of lines problem, where the input is a set $L$ of lines in $\mathbb{R}^d$, and the goal is to find $k$ centers $C$ in $\mathbb{R}^d$ such that the sum of squared distances from each line in $L$ to its nearest center in $C$ is minimized. The local search algorithm is a well-established strategy for point-based $k$-means clustering, known for its efficiency and provable approximation guarantees. However, extending local search algorithm to the $k$-means of lines problem is nontrivial, as the capture relation used in point-based clustering does not generalize to the line setting. This is because that the point-to-line distance function lack the triangle inequality property that supports geometric analysis in point-based clustering. Moreover, since lines extend infinitely in space, it is difficult to identify effective swap points that can significantly reduce the clustering cost. To overcome above obstacles, we introduce a *proportional capture relation* that links optimal and current centers based the assignment proportions of lines, enabling a refined analysis that bypasses the triangle inequality barrier. We also introduce a *CrossLine* structure, which provides a principled discretization of the geometric space around line pairs, and ensures coverage of high-quality swap points essential for local search, thereby enabling effective execution of the local search process. Consequently, based on the proposed components, we develop the first single-swap local search algorithm for the $k$-means of lines problem, achieving a $(500+\varepsilon)$-approximation in polynomial time for low-dimensional Euclidean space.

Graph self-supervised learning (GSL) is essential for processing graph-structured data, reducing the need for manual labeling. Traditionally, this paradigm has extensively utilized Bayesian Personalized Ranking (BPR) as its primary loss function. Despite its widespread application, the theoretical analysis of its node relations evaluation have remained largely unexplored. This paper employs recent advancements in latent hyperbolic geometry to deepen our understanding of node relationships from a graph-theoretical perspective. We analyze BPR’s limitations, particularly its reliance on local connectivity through 2-hop paths, which overlooks global connectivity and the broader topological structure. To address these shortcomings, we purpose a novel loss function, BPR+, designed to encompass even-hop paths and better capture global connectivity and topological nuances. This approach facilitates a more detailed measurement of user-item relationships and improves the granularity of relationship assessments. We validate BPR+ through extensive empirical testing across five real-world datasets and demonstrate its efficacy in refining graph self-supervised learning frameworks. Additionally, we explore the application of BPR+ in drug repositioning, highlighting its potential to support pharmaceutical research and development. Our findings not only illuminate the success factors of previous methodologies but also offer new theoretical insights into this learning paradigm.

Alzheimer's disease (AD) is a complex disease with its genetic etiology not fully understood. Gene network-based methods have been proven promising in predicting AD genes. However, existing approaches are limited in their ability to model the nonlinear relationship between networks and disease genes, because (i) any data can be theoretically decomposed into the sum of a linear part and a nonlinear part, (ii) the linear part can be best modeled by a linear model since a nonlinear model is biased and can be easily overfit, and (iii) existing methods do not separate the linear part from the nonlinear part when building the disease gene prediction model. To address the limitation, we propose linear model-integrated graph convolutional network (LIMO-GCN), a generic disease gene prediction method that models the data linearity and nonlinearity by integrating a linear model with GCN. The reason to use GCN is that it is by design naturally suitable to dealing with network data, and the reason to integrate a linear model is that the linearity in the data can be best modeled by a linear model. The weighted sum of the prediction of the two components is used as the final prediction of LIMO-GCN. Then, we apply LIMO-GCN to the prediction of AD genes. LIMO-GCN outperforms the state-of-the-art approaches including GCN, network-wide association studies, and random walk. Furthermore, we show that the top-ranked genes are significantly associated with AD based on molecular evidence from heterogeneous genomic data. Our results indicate that LIMO-GCN provides a novel method for prioritizing AD genes.

MOTIVATION: DNA methylation plays important roles in various cellular physiological processes in bacteria. Nanopore sequencing has shown the ability to identify different types of DNA methylation from individual bacteria directly. However, existing methods for identifying bacterial methylomes showed inconsistent performances in different methylation motifs in bacteria and didn't fully utilize the different scale information contained in nanopore signals. RESULTS: We propose a deep-learning method, called Nanoident, for de novo detection of DNA methylation types and methylated base positions in bacteria using Nanopore sequencing. For each targeted motif sequence, Nanoident utilizes five different features, including statistical features extracted from both the nanopore raw signals and the basecalling results of the motif. All the five features are processed by a multi-scale neural network in Nanoident, which extracts information from different receptive fields of the features. The Leave-One-Out Cross-Validation (LOOCV) on the dataset containing 7 bacteria samples with 46 methylation motifs shows that Nanoident achieves ∼10% improvement in accuracy than the previous method. Furthermore, Nanoident achieves ∼13% improvement in accuracy in an independent dataset, which contains 12 methylation motifs. Additionally, we optimize the pipeline for de novo methylation motif enrichment, enabling the discovery of novel methylation motifs. AVAILABILITY AND IMPLEMENTATION: The source code of Nanoident is freely available at https://github.com/cz-csu/Nanoident and https://doi.org/10.6084/m9.figshare.29252264.

MOTIVATION: Nanopore sequencing data offer longer reads compared to other technologies, which is beneficial for phasing and genome assembly. INDELs provide valuable haplotype information and have significant potential to improve phasing performance. However, accurately identifying INDELs with variant callers is challenging, and incorporating INDELs into phasing remains a complex task. To address these issues, we developed NanoFilter, a novel filtering strategy designed to filter out INDELs that contain wrong phasing information based on their consistency. RESULTS: Our assessment using Nanopore R10 simplex data shows that filtering out low-consistency INDELs increases their precision from 88.3% to 98.8%, nearly matching the precision of SNVs. In the phasing results of Margin, incorporating these filtered INDELs leads to a 12.77% increase in N50 length and fewer switch errors. Furthermore, we found that SNVs filtered by NanoFilter will enhance assembly performance. When NanoFilter is integrated into the HapDup assembly pipeline, NanoFilter reduces the Hamming error rate and increases N50 length by 7.8%. AVAILABILITY AND IMPLEMENTATION: NanoFilter is available at https://github.com/Chenshanming-repo/NanoFilter (DOI: 10.5281/zenodo.16777826) and HapDup-NanoFilter is available at https://github.com/Chenshanming-repo/HapDup-NanoFilter (DOI: 10.5281/zenodo.16777890).

MOTIVATION: Drug combination therapy is an effective strategy for cancer treatment, enhancing drug efficacy and reducing toxic side effects. However, in vitro drug screening experiments are time-consuming and expensive, necessitating the development of computational methods for drug synergy prediction. While current methods focus on molecular chemical structures, they often overlook the biological context, limiting their ability to capture complex drug synergies. RESULTS: In this work, we propose MADSP, a novel method for anti-cancer drug synergy prediction that integrates target and pathway knowledge for a more comprehensive understanding of systems biology. MADSP first incorporates chemical structure, target, and pathway features of drugs, using a multi-head self-attention mechanism to learn a unified drug representation. It then integrates protein-protein interaction data with omics data from cell lines, extracting a low-dimensional dense embedding of cell lines via an autoencoder. Finally, the synergy scores for drug combinations are predicted using a fully connected neural network. Experiments on benchmark datasets demonstrate that MADSP outperforms state-of-the-art methods. The ablation study reveals that multi-source information fusion and attention mechanisms significantly enhance model performance. The case study further illustrates the practical applicability of MADSP as a powerful tool for drug synergy prediction, offering potential for advancing cancer treatment strategies. AVAILABILITY AND IMPLEMENTATION: MADSP is available at https://github.com/Hhyqi/MADSP.

MOTIVATION: Isoforms spliced from the same gene may carry distinct biological functions. Therefore, annotating functions at the isoform level provides valuable insights into the functional diversity of genomes. Since experimental approaches for determining isoform functions are time- and cost-demanding, computational methods have been proposed. In this case, multi-omics data integration helps enhance the model performance, providing complementary insights for isoform functions. However, current methods underperform in leveraging diverse omics data, primarily due to the limited power to integrate the heterogeneous feature domains. Besides, among the multi-omics data, isoform-isoform interactions (IIIs) are a key data source, as isoforms interact with each other to perform functions. Unfortunately, IIIs remain largely underutilized in isoform function predictions until now. RESULTS: We introduce CrossIsoFun, a multi-omics data analysis framework for isoform function prediction. CrossIsoFun combines omics-specific and cross-omics learning for data integration and function prediction. In detail, CrossIsoFun uses a graph convolutional network (GCN) as the omics-specific classifier for each data source. The initial label predictions from GCNs are forwarded to the View Correlation Discovery Network (VCDN) and processed as a cross-omics integrative representation. The representation is then used to produce final predictions of isoform functions. In addition, an antoencoder within a cycle-consistency generative adversarial network (cycleGAN) is designed to generate IIIs from PPIs and thereby enrich the interactomics data. Our method outperforms the state-of-the-art methods on three tissue-naive datasets and 15 tissue-specific datasets with mRNA expression, sequence, and PPI data. The prediction of CrossIsoFun is further validated by its consistency with subcellular localization and isoform-level annotations with literature support. AVAILABILITY AND IMPLEMENTATION: CrossIsoFun is freely available at https://github.com/genemine/CrossIsoFun.

MOTIVATION: Accurately identifying and predicting molecular properties is a crucial task in molecular machine learning, and the key lies in how to extract effective molecular representations. Contrastive learning opens new avenues for representation learning, and a large amount of unlabeled data enables the model to generalize to the huge chemical space. However, existing contrastive learning-based models face two challenges: (i) existing methods destroy the original molecular environment and ignore chemical prior information, and (ii) there is a lack of a prior knowledge to guide the prediction of molecular properties. RESULTS: In this work, we propose a molecular property prediction framework called MolFCL, which consists of fragment-based contrastive learning and functional group-based prompt learning. Specifically, we introduced fragment-fragment interactions for the first time in the contrastive learning framework and designed a fragment-based augmented molecular graph that integrates the original chemical environment and fragment reactions. Furthermore, we proposed a novel functional group-based prompt learning during fine-tuning, which first incorporates functional group knowledge and the corresponding atomic signals, to improve molecular representation and provide interpretable analyses. The results show that MolFCL outperforms state-of-the-art baseline models on 23 molecular property prediction datasets. Moreover, visualizations show that MolFCL can learn to embed molecules into representations that can distinguish chemical properties. MolFCL can give higher weight to functional groups consistent with chemical knowledge during the prediction of molecular properties, which offers an interpretable ability of the model. Overall, MolFCL is a practically useful tool for molecular property prediction and assists drug scientists in designing drugs more effectively. AVAILABILITY AND IMPLEMENTATION: MolFCL is available at https://github.com/tangxiangcsu/MolFCLSupplementary.

MOTIVATION: Protein-protein interactions (PPIs) are fundamental aspects in understanding biological processes. Accurately predicting the effects of mutations on PPIs remains a critical requirement for drug design and disease mechanistic studies. Recently, deep learning models using protein 3D structures have become predominant for predicting mutation effects. However, significant challenges remain in practical applications, in part due to the considerable disparity in generalization capabilities between easy and hard mutations. Specifically, a hard mutation is defined as one with its maximum TM-score <0.6 when compared to the training set. Additionally, compared to physics-based approaches, deep learning models may overestimate performance due to potential data leakage. RESULTS: We propose new training/test splits that mitigate data leakage according to the CATH homologous superfamily. Under the constraints of physical energy, protein 3D structures, and CATH domain objectives, we employ a hybrid noise strategy as data augmentation and present a geometric encoder scenario, named CATH-ddG, to represent the mutational microenvironment differences between wild-type and mutated protein complexes. Additionally, we fine-tune ESM2 representations by incorporating a lightweight nonlinear module to achieve the transferability of sequence co-evolutionary information. Finally, our study demonstrates that CATH-ddG framework provides enhanced generalization by outperforming other baselines on non-superfamily leakage splits, which plays a crucial role in exploring robust mutation effect regression prediction. Independent case studies demonstrate successful enhancement of binding affinity on 419 antibody variants to human epidermal growth factor receptor 2 (HER2) and 285 variants in the receptor-binding domain (RBD) of SARS-CoV-2 to angiotensin-converting enzyme 2 (ACE2) receptor. AVAILABILITY AND IMPLEMENTATION: CATH-ddG is available at https://github.com/ak422/CATH-ddG.

MOTIVATION: Drug side effects refer to harmful or adverse reactions that occur during drug use, unrelated to the therapeutic purpose. A core issue in drug side effect prediction is determining the frequency of these drug side effects in the population, which can guide patient medication use and drug development. Many computational methods have been developed to predict the frequency of drug side effects as an alternative to clinical trials. However, existing methods typically build regression models on five frequency classes of drug side effects and tend to overfit the training set, leading to boundary handling issues and the risk of overfitting. RESULTS: To address this problem, we develop a multi-source similarity fusion-based model, named multi-source similarity fusion (MSSF), for predicting five frequency classes of drug side effects. Compared to existing methods, our model utilizes the multi-source feature fusion module and the self-attention mechanism to explore the relationships between drugs and side effects deeply and employs Bayesian variational inference to more accurately predict the frequency classes of drug side effects. The experimental results indicate that MSSF consistently achieves superior performance compared to existing models across multiple evaluation settings, including cross-validation, cold-start experiments, and independent testing. The visual analysis and case studies further demonstrate MSSF's reliable feature extraction capability and promise in predicting the frequency classes of drug side effects. AVAILABILITY AND IMPLEMENTATION: The source code of MSSF is available on GitHub (https://github.com/dingxlcse/MSSF.git) and archived on Zenodo (DOI: 10.5281/zenodo.15462041).

MiRNAs play an important role in the occurrence and development of human disease. Identifying potential miRNA-disease associations is valuable for disease diagnosis and treatment. Therefore, it is urgent to develop efficient computational methods for predicting potential miRNA-disease associations to reduce the cost and time associated with biological wet experiments. In addition, high-quality feature representation remains a challenge for miRNA-disease association prediction using graph neural network methods. In this paper, we propose a method named ESGC-MDA, which employs an enhanced Simple Graph Convolution Network to identify miRNA-disease associations. We first construct a bipartite attributed graph for miRNAs and diseases by computing multi-source similarity. Then, we enhance the feature representations of miRNA and disease nodes by applying two strategies in the simple convolution network, which include randomly dropping messages during propagation to ensure the model learns more reliable feature representations, and using adaptive weighting to aggregate features from different layers. Finally, we calculate the prediction scores of miRNA-disease pairs by using a fully connected neural network decoder. We conduct 5-fold cross-validation and 10-fold cross-validation on HDMM v2.0 and HMDD v3.2, respectively, and ESGC-MDA achieves better performance than state-of-the-art baseline methods. The case studies for cardiovascular disease, lung cancer and colon cancer also further confirm the effectiveness of ESGC-MDA.

The excessive use of antibiotics has contributed to the growing resistance of various microbial pathogens, prompting the exploration of alternatives such as antimicrobial peptides (AMPs). However, wet-lab experimental methods for AMP identification are not only time-consuming but also resource-intensive. Machine learning (ML) and deep learning (DL) techniques have been increasingly employed for AMP classification to overcome these limitations. Recently, pre-trained protein language models (pLMs) have emerged, demonstrating their exceptional ability to represent biological sequences in a variety of prediction tasks. Building on these advancements, we present KT-AMPpred, a method for predicting antimicrobial peptides and their specific antimicrobial properties using transfer learning and fine-tuning techniques based on pre-trained protein language models. Utilizing transfer learning, KT-AMPpred effectively integrates knowledge from prior AMP classification tasks to improve the prediction of distinct antimicrobial properties. Additionally, the fine-tuning of pLM within KT-AMPpred further enhances its performance on downstream prediction tasks. The experimental outcomes reveal that KT-AMPpred exceeds current leading approaches on benchmark datasets, with visual analysis confirming its robust feature extraction capabilities, establishing it as an effective tool for antimicrobial peptide prediction.

MicroRNAs (miRNAs) play a significant role in biological processes, serving as potential biomarkers and therapeutic targets for disease diagnosis and treatment. However, traditional experimental methods for identifying miRNA-disease associations (MDAs) are costly and time-consuming. These challenges prompt a critical need for computational approaches. This study proposes GCNDFMDA, a method that employs a variant of the deep forest model for MDAs prediction. Unlike simple multi-source integration, GCNDFMDA first captures the interactive features of miRNAs and diseases by extracting diverse information sources for comprehensive representation. Next, it utilizes graph convolutional networks, followed by neural projection, to derive latent representations from similarity networks, thereby reducing dimensionality and obtaining an optimal feature space. It then calculates integrated similarities for miRNAs and diseases and constructs feature vectors. Finally, it employs a variant of the deep forest model to obtain the final prediction. Experimental results on three datasets (HMDD v2.0, HMDD v3.2, and an independent dataset), along with de novo validation, highlight the superior performance of GCNDFMDA compared to seven compared methods. In addition, case studies confirm its reliability, with all the top 50 predicted miRNAs related to these diseases verified in real applications.

Molecular property prediction holds significant importance in the fields of cheminformatics and drug discovery. Current modeling paradigms used for molecular representation mainly rely on 1D or 2D molecular formats, which are unable to distinguish between common stereoisomers, especially conformational and chiral isomers with the same bond connections but different spatial configurations. In addition, a single molecular representation paradigm inhibits the versatility and adaptability of models in various modes. To address these challenges, we propose a Multi-Graph Representation Fusion Network (MGRFN) which employs Graph Attention Network and SphereNet to extract 2D chemical features and 3D geometric information respectively, and design a bilinear fusion module to achieve efficient integration of multimodal representations. Experimental results on the QM9, MD17, and two chiral molecular datasets demonstrate the superior performance of MGRFN in predicting molecular quantum chemical properties and various conformational properties. Moreover, the visualization of molecular representations and attention weights shows that MGRFN can distinguish the physicochemical properties of different molecules and capture molecule-related substructures, which further enhances our understanding of its performance.

Drug-target affinity (DTA) prediction plays a crucial role in accelerating the drug development process. Although deep learning-based models achieve strong performance in benchmark datasets, their predictive accuracy declines sharply in cold-start scenarios, i.e., when encountering drugs or proteins absent from the training set. This limitation arises from the restricted scale of training datasets, leading to features learned by end-to-end DTA models lacking sufficient generalization. To address this challenge, we propose a novel approach named LLMDTA (Large Language Model for DTA), leveraging the power of biological language models to tackle the cold-start problem in DTA prediction. Specifically, we use Mol2Vec, a molecular pre-training model, and ESM2, a protein language model, as feature extractors. To seamlessly integrate these pre-trained features into downstream DTA prediction, we employ a 1D-CNN-based encoder to extract independent molecular features. In addition, a bilinear attention module is designed to capture interactive molecular features between drugs and proteins. Finally, independent and interactive features are fused to predict binding affinities. Experimental results in three benchmark datasets demonstrate that LLMDTA consistently outperforms state-of-the-art baselines in both warm-start and cold-start scenarios, with notable improvements in novel-protein and novel-pair settings. Furthermore, a case study involving the epidermal growth factor receptor illustrates the ability of LLMDTA to identify novel binding affinities between previously unseen drugs and the target protein, validated through molecular docking. In general, LLMDTA represents a promising and practical tool for advancing DTA prediction in real-world applications.

Probiotics are active microorganisms that provide substantial health benefits. They can serve as an alternative or complement to medications, enabling effective and targeted treatment for various diseases. However, traditional experimental methods for screening probiotics are time-consuming and labor-intensive, underscoring the need for efficient computational approaches. While some studies have introduced link prediction methods based on known probiotic-disease associations, these methods often fail to address errors and noise within the dataset and overlook the rich, intrinsic features of probiotics and diseases. To address these limitations, this paper presents MFFPDA, the first deep-learning framework based on multi-feature fusion for predicting probiotic-disease associations. We systematically screened a probiotic-disease association dataset and collected probiotic and disease-related data from various sources. Furthermore, we calculate multiple features of probiotics and diseases, and design some feature extraction and fusion modules to integrate these features. The comparison results demonstrate that MFFPDA surpasses all other comparison methods. Feature visualization results confirmed the necessity and rationale for incorporating multi-source features. Additionally, case studies on colonic pseudo-obstruction and dysentery further validated the effectiveness of MFFPDA, underscoring its reliability as a tool for predicting probiotic-disease associations.

Molecular associations are central to most biological processes. The discovery and identification of potential associations between molecules can provide insights into biological exploration, diagnostic and therapeutic interventions, and drug development. So far many relevant computational methods have been proposed, but most of them are usually limited to specific domains and rely on complex preprocessing procedures, which restricts the models' ability to be applied to other tasks. Therefore, it remains a challenge to explore a generalized approach to accurately predicting potential associations. In this study, We propose Left-Right Transition Matrices (LRTM) for molecular association prediction. From the perspective on the diffusion model, we construct two transition matrices to model undirected graph information propagation. This allows modeling the transition probabilities of links, which facilitates link prediction in molecular bipartite networks. The extensive experimental results show that the proposed LRTM algorithm performs better than the compared methods. Also, the proposed algorithm has the potential for cross-task prediction. Furthermore, case studies show that LRTM is a powerful tool that can be effectively applied to practical applications.

Federated learning (FL) serves as a distributed framework facilitating privacy-preserving machine learning on vast data generated by mobile and IoT devices. Hierarchical federated averaging (H-FedAvg) has emerged as a solution to the communication bottleneck, enabling large-scale implementation. It leverages the client-edge-cloud aggregation hierarchy, capitalizing on the accessibility of the cloud server and the communication efficiency of edge servers (ESs). However, the full potential of H-FedAvg in real-world scenarios, particularly in the presence of imbalanced ES communication latency, remains largely unexplored. This article delves into the issue of imbalanced communication latency and introduces the edge-level imbalanced H-FedAvg (IH-FedAvg) algorithm, accompanied by convergence analysis. Theoretical and empirical results underscore the degradation of IH-FedAvg, which is influenced by the number of edge aggregation rounds and the divergence between the slowest and fastest ESs. To tackle this challenge, we propose an error-feedback compression method, leading to the edge-level compressed H-FedAvg (ECH-FedAvg) algorithm. Theoretical analysis indicates that the convergence performance depends on the compressor ratio. Going beyond compression, we present a joint optimization approach that considers both compressor ratio selection and bandwidth allocation, dynamically guiding ECH-FedAvg. Extensive experiments show that ECH-FedAvg achieves competitive performance compared to H-FedAvg, significantly enhancing communication efficiency. These findings underscore the potential of ECH-FedAvg in addressing challenges related to imbalanced communication in real-world networks.

To meet the stringent requirements of industrial applications, modern Ethernet datacenter networks widely deployed with remote direct memory access (RDMA) technology and priority-based flow control (PFC) scheme aim at providing low latency and high throughput transmission performance. However, the existing end-to-end congestion control cannot handle the transient congestion timely due to the round-trip-time (RTT) level control loop, inevitably resulting in PFC triggering. In this article, we propose a Sub-RTT congestion control mechanism called SRCC to alleviate bursty congestion timely. Specifically, SRCC identifies the congested flows accurately, notifies congestion directly from the hotspot to the corresponding source at the sub-RTT control loop and adjusts the sending rate to avoid PFC's head-of-line blocking. Compared to the state-of-the-art end-to-end transmission protocols, the evaluation results show that SRCC effectively reduces the average flow completion time (FCT) by up to 61%, 52%, 40%, and 24% over datacenter quantized congestion notification (DCQCN), Swift, high precision congestion control (HPCC), and photonic congestion notification (PCN), respectively.

Accurate patient outcome predictions are essential for healthcare improvement, yet utilizing diverse medical data raises privacy and security concerns. Federated learning enables collaborative model training while preserving data privacy. However, the heterogeneity of clinical features among hospitals poses a challenge, leading to suboptimal performance in conventional federated learning. In response, we propose FedComDist, a personalized federated learning approach designed to maximize the use of heterogeneous features across hospitals for patient outcome prediction. Our approach incorporates a novel method for optimizing common global and distinct local parameters. We categorize input clinical features into two main groups-common and distinct-based on their presence across all hospitals and decouple the model parameters into common and distinct accordingly. The common features are used to train common global parameters, which are aggregated and optimized on the server, making them trainable across all hospitals. Meanwhile, the distinct features are used to train local parameters and optimized using the local dataset of each hospital. Our approach is evaluated on the eICU dataset, a publicly available multi-center clinical dataset, to predict patient clinical outcomes, specifically mortality and Length of Stay (LoS). The experimental results demonstrate the effectiveness of our approach compared to various federated learning methods and provide enhanced privacy through parameter decoupling.

The electrocardiogram (ECG) serves as a crucial tool for myocardial infarction (MI) localization, and deep learning methods have proven effective in assisting physicians with MI localization. Traditional MI localization methods are purely data-driven, and the quality of the data significantly affects the model's performance, particularly in the localization of rare MI. We propose a knowledge-driven graph representation learning (KD-GRL) framework which is designed to guide deep learning models in identifying key features for MI localization using prior knowledge. The MI localization knowledge graph (KG) is constructed by integrating medical knowledge about MI localization, including ECG leads and morphological manifestations, the correlations between MI localization labels, diagnostic rules, and patient demographic information. KG effectively represents the relationships among various entities, which include ECG signal entities, morphological feature entities, and demographic feature entities. The embeddings of these entities are obtained using parallel patient multi-feature extractors. Additionally, a KG aggregation method based on edge relation projection (ERP) is proposed to aggregate the relational information in the MI localization KG. Ultimately, the MI localization task is transformed into a link prediction task between patient entity and localization label entities within the KG. We conduct experiments on two public datasets, PTB and PTBXL, achieving F1-scores of 48.90% and 46.06%, respectively, both surpassing the comparison methods. Additionally, due to the incorporation of diagnostic knowledge, our method outperforms the comparison methods in localizing rare MIs.

Pathological diagnosis assists in saving human lives, but such models are annotation hungry and pathological images are notably expensive to annotate. Contrastive learning could be a promising solution that relies only on the unlabeled training data to generate informative representations. However, the majority of current methods in contrastive learning have the following two issues: (1) positive samples produced through random augmentation are less challenging, and (2) false negative pairs problem caused by negative sampling bias. To alleviate the above issues, we propose a novel contrastive learning method called Cluster-Aware Adversarial Contrastive Learning (CA2CL). Specifically, a mixed data augmentation technique is provided to learn more transferable representations by generating more discriminative sample pairs. Furthermore, to mitigate the effects of inherent false negative pairs, we adopt a cluster-aware loss to identify similarities between instances and incorporate them into the process of contrastive learning. Finally, we generate challenging contrastive data pairs by adversarial learning, and adversarially learn robust representations in the representation space without the labeled training data, which aims to maximize the similarity between the augmented sample and the related adversarial sample. Our proposed CA2CL is evaluated on two public datasets: NCT-CRC-HE and PCam for the fine-tuning and linear evaluation tasks and on two other public datasets: GlaS and CARG for the detection and segmentation tasks, respectively. Extensive experimental results demonstrate the superior performance improvement of our method over several Self-supervised learning (SSL) methods and ImageNet pretraining particularly in scenarios with limited data availability for all four tasks.

Anticancer drug response prediction is crucial in developing personalized treatment plans for cancer patients. However, High-quality patient anticancer drug response data are scarce and cell line data and patient data have different distributions, models trained solely on cell line data perform poorly. Some existing methods predict anticancer drug response by transferring knowledge from the cell line domain to the patient domain using transfer learning. However, the robustness of these classifiers is affected by anomalies in the cell line data, and they do not utilize the knowledge in the unlabeled target domain data. To this end, we proposed a model called DAPL to predict patient responses to anticancer drugs. The model extracts domain-invariant features from cell lines and patients by constructing multiple VAEs and extracts drug features using GNNs. These features are then combined for prototypical learning to train a classifier, resulting in better predictions of patient anticancer drug response. We used the cell line datasets CCLE and GDSC as source domains and the patient datasets TCGA and PDTC as target domains and conducted experiments. The results indicate that DAPL shows excellent performance in predicting patient anticancer drug response compared to other state-of-the-art methods.

Patients with the same type of cancer often respond differently to identical drug treatments due to unique genomic traits. Accurately predicting a patient's response to drug is crucial in guiding treatment decisions, alleviating patient suffering, and improving cancer prognosis. Current computational methods utilize deep learning models trained on extensive drug screening data to predict anti-cancer drug responses based on features of cell lines and drugs. However, the interaction between cell lines and drugs is a complex biological process involving interactions across various levels, from internal cellular and drug structures to the external interactions among different molecules.To address this complexity, we propose a novel Hierarchical graph representation Learning with Multi-Granularity features (HLMG) algorithm for predicting anti-cancer drug responses. The HLMG algorithm combines features at two granularities: the overall gene expression and pathway substructures of cell lines, and the overall molecular fingerprints and substructures of drugs. Subsequently, it constructs a heterogeneous graph including cell lines, drugs, known cell line-drug responses, and the associations between similar cell lines and similar drugs. Through a graph convolutional network model, the HLMG learns the final cell line and drug representations by aggregating features of their multi-level neighbor in the heterogeneous graph. The multi-level neighbors consist of the node self, directly related drugs/cell lines, and indirectly related similar drugs/cell lines. Finally, a linear correlation coefficient decoder is employed to reconstruct the cell line-drug correlation matrix to predict anti-cancer drug responses. Our model was tested on the Genomics of Drug Sensitivity in Cancer (GDSC) and the Cancer Cell Line Encyclopedia (CCLE) databases. Results indicate that HLMG outperforms other state-of-the-art methods in accurately predicting anti-cancer drug responses.

Self-supervised learning on graph neural networks is receiving increasing attention due to the difficulty of obtaining graph labels in many real applications. Graph contrastive learning (GCL), a recently popular method for self-supervised learning on graphs, has achieved great success in many tasks. The key to the effectiveness of GCL is the construction of suitable contrasting pairs to capture important attributes of the data through the data augmentation modules. However, most of the existing approaches fail to fully consider both data diversity and the semantic consistency when conducting data augmentation. To fill this gap, we propose an augmentation diversity and semantic consistency balanced graph contrastive learning model (ADSCB for short), which enhances the representation ability of the CL model through richer contrasting objectives. In particular, we first introduce a semantic consistency module to extract the subgraph from the original graph through optimizing a carefully designed semantic consistency loss. Then, we introduce an augmentation diversity module and perform data augmentation and cross-scale mix-up operations on the original graph and the extracted semantic preserved subgraph to generate more diverse contrasting pairs. With the above two modules, our model ultimately achieves two contrasting objectives: diversity contrasting and semantic contrasting. The tradeoff between these two contrasting objectives allows our model to benefit from both the augmentation diversity and the semantic consistency. We evaluate ADSCB for graph classification in unsupervised, semi-supervised, and transfer learning settings using standard graph contrastive learning benchmarks. The results demonstrate the superiority of our method against several state-of-the-art baselines.

To meet the stringent demanding low latency and high throughput of cloud datacenter applications, recent receiver-driven transport protocols transmit only one packet once receiving each credit packet from the receiver to achieve ultra-low queueing delay. However, the round-trip time variation and the highly dynamic background traffic significantly deteriorate the performance of receiver-driven transport protocols, resulting in under-utilized bandwidth. This paper designs a simple yet effective solution called RPO, which retains the advantages of receiver-driven transmission while efficiently utilizing the available bandwidth. Specifically, RPO rationally uses low-priority opportunistic packets to ensure high network utilization without increasing the queueing delay of high-priority normal packets. Furthermore, to tackle the queueing buildup due to line-rate transmission in the first RTT, we design a selective dropping mechanism called SDM to help the majority of small flows complete within only one RTT by prioritizing the first-RTT bursty packets over the packets triggered by grants. We implement RPO in Linux hosts with DPDK. The experimental results show that RPO significantly improves the network utilization by up to 35% over the state-of-the-art schemes, without introducing additional queueing delay. Moreover, RPO integrated with SDM reduces the AFCT of small flows by up to 45% compared with RPO integrated with Aeolus.

Nowadays, Congestion Control (CC) algorithms are expected to satisfy the diverse demands of applications running over diverse networks. To achieve this goal, the combinations, which are expected to inherit both the advantages of classic CC in terms of convergence, overhead, and explainability, and the advantages of learning-based CC on adapting to diverse networks and demands, become a hot topic. In this paper, we reveal the existing combination works are eithergiving a fishorteaching to fish. Based on the insight of their essential issues, we develop the Concentrator method ofteaching to fish. According to this method, we propose Seagull as a step further. Specifically, Seagull captures the network characteristics and application demands in a coarse-grained manner via an online learning module. Moreover, the online learning module guides the customization of the rate adjustment rules of the classic CC module for fine-grained system evolution. Replacing the assumption on networks by the captured characteristics, the classic CC module of Seagull can fulfill the specified application demands. Real-world experimental results show Seagull respectively outperforms Orca, PCC-Vivace, and CUBIC by$49.3\%,\ 30.4\%$, and 24.9% in terms of throughput over the Internet, and improves the video quality of experience (QoE) by$12.9\sim 33.5\%$compared to CUBIC over cellular links.

With the proliferation of mobile sensing techniques, huge amounts of time series data are continuously generated and accumulated in various domains, fueling considerable real-world mobile computing applications. In this context, time series anomaly detection is practically important. It endeavors to identify deviant samples from the normal distribution in time series data. Existing approaches generally assume that all the time series data is available at a central location. However, with the increasing deployment of edge devices, we are witnessing a decentralized collection of time series data. To bridge the gap between decentralized data and centralized anomaly detection algorithms, we propose a Parameter-efficient Federated Anomaly Detection framework via Lightweight Adaptation (PeFAD-LA) that addresses growing privacy concerns. PeFAD-LA innovatively employs a pre-trained large language model (PLM or LLM) as the core of the client's local model, which can benefit from its cross-modality knowledge transfer capability. To reduce the communication overhead and the local model adaptation cost, we propose a parameter-efficient federated training module that requires clients to fine-tune only small-scale parameters and transmit them to the server for updates. Further, to handle anomalies in streaming time series, a lightweight adaptation module is employed to overcome concept drift. PeFAD-LA utilizes an anomaly-driven mask selection strategy to mitigate the impact of neglected anomalies during training. A novel dual knowledge transfer mechanism is designed to harness the useful knowledge across clients and sequentially learned data. Extensive experiments on real data offer evidence of the effectiveness and efficiency of the proposed framework.

Large-scale machine learning typically adopts distributed machine learning (DML) techniques to accelerate model training. Due to the large communication overhead, unfortunately, the phase of gradient aggregation has become the performance bottleneck for data-parallel DML. To reduce traffic volume, several in-network aggregation (INA) transmission protocols are proposed to offload gradient aggregation function into the programmable switches. However, since existing INA transmission protocols use synchronous congestion control mechanism to drive each round of gradient aggregation, the straggling workers lead to long iteration time and significant performance degradation. Besides, we reveal that existing INA solutions cannot provide the fairness performance among multiple jobs with varying number of workers. To solve the above problem, we propose PA-ATP, a progress-aware INA transmission protocol, which adopts the progress-aware asynchronous congestion control. PA-ATP adjusts the sending rate in accordance with the transmission progress, allowing the straggling flow to grab more bandwidth than the leading flow and control the asynchronous degree of straggling job. Moreover, to ensure the fair throughput among multiple jobs, we dynamically adjust the aggregator allocation for each job by tuning the number of hash operations. We use a P4 programmable switch and a kernel-bypass protocol stack to implement PA-ATP. The results of testbed and large-scale NS3 simulations show that PA-ATP reduces training time by up to 62% compared to the state-of-the-art INA transmission protocols.

For the widely deployed on-chip shared buffer, efficient buffer management is the key to absorbing bursts and avoiding packet loss during transient congestion. However, as the buffer-per-port-per-Gbps in production data centers decreases, it becomes more challenging to provide efficient buffer management to meet the requirements of heterogeneous traffic. We observe that typical shared buffer management policies have two steps: first, they identify short flows arriving at ports and then allocate more buffer room for these ports. Unfortunately, the lack of isolation between long and short flows leads to increased queue buildup and even packet loss of short flows. To address this limitation, we propose D2T, which uses different queue length thresholds for long and short flows. Specifically, we first design a compact data structure to distinguish between long and short flows. Then when two kinds of flows coexist at the same port, the threshold of long flows will decrease to absorb the bursty short flows. What’s more, we introduce D2T${}^{*}$which combines D2T with advanced DRL techniques to move toward mastering buffer management for further improving performance across various scenarios. We implement D2T at a P4-programmable switch and large-scale simulations. The results demonstrate that D2T reduces both average and tail flow completion times (FCT) of short flows by up to 29% and 62% compared with the state-of-the-art policies, respectively.

As one of the most popular machine learning tools in the field of unsupervised learning, clustering has been widely used in various practical applications. While numerous methods have been proposed for clustering, a commonly encountered issue is that the existing clustering methods rely heavily on local neighborhood information during the optimization process, which leads to suboptimal performance on real-world datasets. Besides, most existing clustering methods use Euclidean distances or densities to measure the similarity between data points. This could constrain the effectiveness of the algorithms for handling datasets with irregular patterns. Thus, a key challenge is how to effectively capture the global structural information in clustering instances to improve the clustering quality. In this paper, we propose a new clustering algorithm, called SEC. This algorithm uses the global structural information extracted from an encoding tree to guide the clustering optimization process. Based on the relation between data points in the instance, a sparse graph of the clustering instance can be constructed. By leveraging the sparse graph constructed, we propose an iterative encoding tree method, where hierarchical abstractions of the encoding tree are iteratively extracted as new clustering features to obtain better clustering results. To avoid the influence of easily misclustered data points located on the boundaries of the clustering partitions, which we call "fringe points", we propose an iterative pre-deletion and reassignment technique such that the algorithm can delete and reassign the "fringe points" to obtain more resilient and precise clustering results. Empirical experiments on both synthetic and real-world datasets demonstrate that our proposed algorithm outperforms state-of-the-art clustering methods and achieves better clustering performances. On average, the clustering accuracy (ACC) is increased by 1.7% and the normalized mutual information (NMI) by 7.9% compared with the current state-of-the-art (SOTA) algorithm on synthetic datasets. On real-world datasets, our method outperforms other clustering methods with an average increase of 12.3% in ACC and 5.2% in NMI, respectively.

Unsupervised rationale extraction aims to extract text snippets to support model predictions without explicit rationale annotation.Researchers have made many efforts to solve this task.Previous works often encode each aspect independently, which may limit their ability to capture meaningful internal correlations between aspects.While there has been significant work on mitigating spurious correlations, our approach focuses on leveraging the beneficial internal correlations to improve multi-aspect rationale extraction.In this paper, we propose a Multi-Aspect Rationale Extractor (MARE) to explain and predict multiple aspects simultaneously.Concretely, we propose a Multi-Aspect Multi-Head Attention (MAMHA) mechanism based on hard deletion to encode multiple text chunks simultaneously.Furthermore, multiple special tokens are prepended in front of the text with each corresponding to one certain aspect.Finally, multi-task training is deployed to reduce the training overhead.Experimental results on two unsupervised rationale extraction benchmarks show that MARE achieves state-ofthe-art performance.Ablation studies further demonstrate the effectiveness of our method.

Nowadays the data center switches employ the on-chip shared buffer to absorb bursts and avoid packet loss during transient congestion. However, as the buffer-per-port-per-Gbps in production data centers decreases, it becomes more challenging to provide efficient buffer management to meet the requirements of heterogeneous traffic. We observe that typical shared buffer management policies have two steps: first, they identify short flows arriving at ports and then allocate more buffer room for these ports. Unfortunately, the lack of isolation between long and short flows leads to increased queue buildup and even packet loss of short flows. To address this limitation, we propose D2T, which uses different queue length thresholds for long and short flows. Specifically, we first design a compact data structure to distinguish between long and short flows. Then when two kinds of flows coexist at the same port, the threshold of long flows will decrease to absorb the bursty short flows. We implement D2T at a P4- programmable switch and large-scale simulations. The results demonstrate that D2T reduces both average and tail flow completion times (FCT) of short flows by up to 29% and 62% compared with the state-of-the-art policies, respectively.

Federated Domain Generalization (FedDG) aims to train the global model for generalization ability to unseen domains with multi-domain training samples. However, clients in federated learning networks are often confined to a single, non-IID domain due to inherent sampling and temporal limitations. The lack of cross-domain interaction and the in-domain divergence impede the learning of domain-common features and limit the effectiveness of existing FedDG, referred to as the single-source FedDG (sFedDG) problem. To address this, we introduce the Federated Global Consistent Augmentation (FedGCA) method, which incorporates a style-complement module to augment data samples with diverse domain styles. To ensure the effective integration of augmented samples, FedGCA employs both global guided semantic consistency and class consistency, mitigating inconsistencies from local semantics within individual clients and classes across multiple clients. The conducted extensive experiments demonstrate the superiority of FedGCA.

Federated Learning (FL) presents a decentralized learning approach for privacy preservation. While many research focuses on uni-modal FL, a more intricate version, multimodal FL, uncovers fundamental attributes. Clients might only gather specific modalities, incurring missing modalities in multi-modal FL. The missing modality problem incurs modality heterogeneity among clients and loses inter-modal connection within one client. To address these issues, we introduce a novel multi-modal FL method called Federated Missing Modality Reconstruction (FedMMR), which tackles these dual challenges through two distinct facets. First, we devise a cross-modal reconstruction policy for synthesizing absent modalities from other modalities. This aligns the data feature spaces across clients, thereby alleviating bias in resultant local models. Subsequently, we steer local models to maintain an inter-modal awareness of both existing and reconstructed modalities, recognizing their potential as complementary components. Comprehensive results show that FedMMR outperforms existing FL baselines.

The k-means with outliers problem is one of the most extensively studied clustering problems in the field of machine learning, where the goal is to discard up to z outliers and identify a minimum k-means clustering on the remaining data points. Most previous results for this problem have running time dependent on the aspect ratio Δ (the ratio between the maximum and the minimum pairwise distances) to achieve fast approximations. To address the issue of aspect ratio dependency on the running time, we propose sampling-based algorithms with almost linear running time in the data size, where a crucial component of our approach is an algorithm called Fast-Sampling. Fast-Sampling algorithm can find inliers that well approximate the optimal clustering centers without relying on a guess for the optimal clustering costs, where a 4-approximate solution can be obtained in time $O(\frac{ndk\log\log n}{\epsilon^2})$ with O(k/ϵ) centers opened and (1+ϵ)z outliers discarded. To reduce the number of centers opened, we propose a center reduction algorithm, where an O(1/ϵ)-approximate solution can be obtained in time $O(\frac{ndk\log \log n}{\epsilon^2} + dpoly(k, \frac{1}{\epsilon})\log(n\Delta))$ with (1+ϵ)z outliers discarded and exactly k centers opened. Empirical experiments suggest that our proposed sampling-based algorithms outperform state-of-the-art algorithms for the k-means with outliers problem.

In modern data centers, many flow scheduling schemes are proposed to accelerate data transfer and improve user experience. However, these schemes assume ideally the prior knowledge of the flow size information, which, unfortunately, is hard to obtain without modifying data center applications. The sketch-based approaches measure the flow size at switch with a compact memory structure, high throughput, and acceptable accuracy loss. However, existing sketches commonly focus on large or specific flows, while most flows in data center networks are small, resulting in missing or overestimated size information about small flows. We propose Strainer Sketch, which enables accurate and fast measurement of small flows with small memory and flexible deployment in a variety of scheduling algorithms. Specifically, Strainer Sketch uses the hierarchical structure to mitigate hash collisions between large and small flows, and the probabilistic counting algorithm to mitigate overestimation due to hash collisions between small flows. Furthermore, we propose a packet scheduling algorithm SW-PIFO, which provides the flow discrimination for a huge number of small flows by using a limited number of queues. Through the testbed experiments and simulations of typical data center applications, we show that our scheme reduces the small flow completion time (FCT) by up to 56.7 <?TeX $\%$?> Math 1 compared with flow scheduling using classic sketches.

Visual Semantic Navigation (VSN) aims at navigating a robot to a given target object in a previously unseen scene. To tackle this task, the robot must learn a nimble navigation policy by utilizing spatial patterns and semantic co-occurrence relations among objects in the scene. Prevailing approaches extract scene priors from the instant visual observations and solidify them in neural episodic memory to achieve flexible navigation. However, due to the oblivion and underuse of the scene priors, these methods are plagued by repeated exploration, effective-knowledge sparsity, and wrong decisions. To alleviate these issues, we propose a novel VSN policy, HSPNav, based on Hierarchical Scene Priors (HSP) and Deep Reinforcement Learning (DRL). The HSP contains two components, i.e., the egocentric semantic map-based Local Scene Priors (LSP) and the commonsense relational graph-based Global Scene Priors (GSP). Then, efficient semantic navigation is achieved by employing an immediate LSP to retrieve conducive contextual memories from the GSP. By utilizing the MP3D dataset, the experimental results in the Habitat simulator demonstrate that our HSP brings a significant boost over the baselines. Furthermore, we take an essential step from simulation to reality by bridging the gap from Habitat to ROS. The migration evaluations show that HSPNav can generalize to realistic settings well and achieve promising performance.

Distributed deep learning has been widely employed to train deep neural network over large-scale dataset. However, the commonly used parameter server architecture suffers from long synchronization time in data-parallel training. Although the existing solutions are proposed to reduce synchronization overhead by breaking the synchronization barriers or limiting the staleness bound, they inevitably experience low convergence efficiency and long synchronization waiting. To address these problems, we propose Gsyn to reduce both synchronization overhead and staleness. Specifically, Gsyn divides workers into multiple groups. The workers in the same group coordinate with each other using the bulk synchronous parallel scheme to achieve high convergence efficiency, and each group communicates with parameter server asynchronously to reduce the synchronization waiting time, consequently increasing the convergence efficiency. Furthermore, we theoretically analyze the optimal number of groups to achieve a good tradeoff between staleness and synchronization waiting. The evaluation test in the realistic cluster with multiple training tasks demonstrates that Gsyn is beneficial and accelerates distributed training by up to 27% over the state-of-the-art solutions.

The strategy for 40/100Gbps Energy Efficient Ethernet (EEE) determines when to enter and leave the power-saving modes. Accordingly, it directly decides both the energy savings and the incurred latency of frames in the EEE. The performance of the EEE strategy is greatly influenced by the network traffic, and thus existing EEE strategies need either proper parameter configuration under certain traffic loads or parameter adaptation mechanisms based on the traffic prediction under the assumptions of certain distribution. Consequently, these EEE strategies hardly keep consistent high performance under variable traffic in reality. To address this issue, we bring the reinforcement learning method into the design of the EEE strategy and propose the reinforcement learning based periodic strategy (RLPS) in this paper. Specifically, RLPS transmits existing frames at first and then stays in the selected power-saving mode for the rest time in each cycle. Moreover, RLPS learns the time length of each cycle online to reflect the impacts of traffic, instead of directly outputting power-saving mode transition decisions. In this way, the power consumption in each cycle is optimal with the help of learned information, and the overhead of online learning is reduced. Simulations driven by both synthetic traffic and real traces confirm that RLPS outperforms existing strategies, i.e., can achieve consistent high performance regardless of the traffic loads and distributions.

Increasingly large-scale models and rich data sets make communication overhead a key bottleneck for distributed Deep Neural Network (DNN) training, constantly attracting the attention of academia and industry. Despite continuous efforts, prior solutions such as pipelining computation/communication and in-network gradient compression/scheduling do not focus on how to accelerate DNN training through load balancing in datacenter networks (DCNs). However, the existing load balancing mechanisms are unaware of tensor integrity and priority for gradient parameter synchronization during the DNN training iterations, resulting in severe tensor tail latency and slow model convergence speed. In this paper, we present a Tensor-aware Load Balancing (TaLB) scheme to accelerate DNN training. Specifically, TaLB identifies the different priority tensors and makes (re)routing decisions based on the tensor-level granularity to cut the high-priority tensors tail delay. The testbed implementation and large-scale NS-3 simulation results show that TaLB effectively accelerates DNN training speed. For example, TaLB significantly reduces the average flow completion time (FCT) by up to 55%, and accelerates the model training speed up to 2.37× on VGG19, ResNet50 and AlexNet models.

Nowadays, Congestion Control (CC) algorithms based on In-Network Telemetry (INT) are popular in datacenter networks, because INT is supported by many commercial devices and provides comprehensive information about network congestion. In this work, we analyze the recent INT-based CC algorithm PowerTCP and reveal its fairness and large delay issues under the conditions of a large number of flows. Inspired by the analytical results, we propose Trident, which enhances PowerTCP by accelerating the speed of converging to fairness and maintaining a low queuing delay with a small equilibrium point. Simulations based on the open source codes confirm that Trident outperforms existing INT-based CC algorithms HPCC, PowerTCP and Poseidon by 14.3%, 14.8%, and 63.5% in terms of FCT of short flows. Moreover, Trident outperforms existing CC algorithms such as Timely, DCQCN and DCTCP, benefiting from the advantages inherited from PowerTCP.

The collective communication has become the bottleneck of large-scale distributed deep learning due to the huge volume of gradients aggregated during the training process. Despite much recent progress in reducing traffic volume by compressing the stochastic gradients inside each training worker, how to share the inter-worker data redundancy to alleviate communication overhead has remained elusive. In this paper, we reveal that most gradients have a great similarity with close value among training workers. From this hypothesis, we propose a Similarity-guided Gradient Compression framework named SGC which skips aggregating the similar gradients among each worker which utilizes local one rather than average value to save communication expenses. Each worker utilizes local SGC firstly quantifies the similarity of gradients among workers, and then elaborately adjusts the aggregation frequency of similar gradients without hurting DNN model accuracy. Meanwhile, we theoretically analyze the convergency accuracy of SGC. The comprehensive evaluation demonstrates that SGC outperforms the state-of-the-art schemes by up to 47% in convergence time.

With the growing ubiquity of video streaming, ensuring a fair and high quality of experience (QoE) for users has emerged as a shared concern among video content providers. State-of-the-art video delivery systems achieve QoE fairness through bottleneck bandwidth allocation across multiple video streaming, all based on the assumption of a unified congestion control (CC) protocol. However, the widespread use of heterogeneous CC protocols on the Internet not only disrupts QoE fairness among video streaming but also poses challenges in achieving fast convergence under dynamic bandwidth. To address these issues, we propose a QoE-Fairness aware bandwidth allocation mechanism called Fabam, which establishes a unified QoE control plane across heterogeneous CC protocols. Fabam constructs independent virtual targets based on the real-time QoE of each video streaming to achieve QoE fairness, and offers rapid convergence for the underlying CC protocols to improve efficiency. We implement Fabam on QUIC and integrate it with Dash.js. The evaluation results demonstrate the significant superiority of Fabam over the state-of-the-art approaches, including an enhancement of 24.48% in QoE fairness and an improvement of 16.63% in QoE efficiency.

With the proliferation of mobile sensing techniques, huge amounts of time series data are generated and accumulated in various domains, fueling plenty of real-world applications. In this setting, time series anomaly detection is practically important. It endeavors to identify deviant samples from the normal sample distribution in time series. Existing approaches generally assume that all the time series is available at a central location. However, we are witnessing the decentralized collection of time series due to the deployment of various edge devices. To bridge the gap between the decentralized time series data and the centralized anomaly detection algorithms, we propose a Parameter-efficient Federated Anomaly Detection framework named PeFAD with the increasing privacy concerns. PeFAD for the first time employs the pre-trained language model (PLM) as the body of the client's local model, which can benefit from its cross-modality knowledge transfer capability. To reduce the communication overhead and local model adaptation cost, we propose a parameter-efficient federated training module such that clients only need to fine-tune small-scale parameters and transmit them to the server for update. PeFAD utilizes a novel anomaly-driven mask selection strategy to mitigate the impact of neglected anomalies during training. A knowledge distillation operation on a synthetic privacy-preserving dataset that is shared by all the clients is also proposed to address the data heterogeneity issue across clients. We conduct extensive evaluations on four real datasets, where PeFAD outperforms existing state-of-the-art baselines by up to 28.74%.

Object Navigation (ObjcetNav), which enables an agent to seek any instance of an object category specified by a semantic label, has shown great advances. However, current agents are built upon occlusion-prone visual observations or compressed 2D semantic maps, which hinder their embodied perception of 3D scene geometry and easily lead to ambiguous object localization and blind exploration. To address these limitations, we present an Embodied Contrastive Learning (ECL) method with Geometric Consistency (GC) and Behavioral Awareness (BA), which motivates agents to actively encode 3D scene layouts and semantic cues. Driven by our embodied exploration strategy, BA is modeled by predicting navigational actions based on multi-frame visual images, as behaviors that cause differences between adjacent visual sensations are crucial for learning correlations among continuous visions. The GC is modeled as the alignment of behavior-aware visual stimulus with 3D semantic shapes by employing unsupervised contrastive learning. The aligned behavior-aware visual features and geometric invariance priors are injected into a modular ObjectNav framework to enhance object recognition and exploration capabilities. As expected, our ECL method performs well on object detection and instance segmentation tasks. Our ObjectNav strategy outperforms state-of-the-art methods on MP3D and Gibson datasets, showing the potential of our ECL in embodied navigation.

The Column Subset Selection (CSS) problem has been widely studied in dimensionality reduction and feature selection. The goal of the CSS problem is to output a submatrix S, consisting of k columns from an n×d input matrix A that minimizes the residual error ‖A-SS^\dagger A‖_F^2, where S^\dagger is the Moore-Penrose inverse matrix of S. Many previous approximation algorithms have non-linear running times in both n and d, while the existing linear-time algorithms have a relatively larger approximation ratios. Additionally, the local search algorithms in existing results for solving the CSS problem are heuristic. To achieve linear running time while maintaining better approximation using a local search strategy, we propose a local search-based approximation algorithm for the CSS problem with exactly k columns selected. A key challenge in achieving linear running time with the local search strategy is how to avoid exhaustive enumerations of candidate columns for constructing swap pairs in each local search step. To address this issue, we propose a two-step mixed sampling method that reduces the number of enumerations for swap pair construction from O(dk) to k in linear time. Although the two-step mixed sampling method reduces the search space of local search strategy, bounding the residual error after swaps is a non-trivial task. To estimate the changes in residual error after swaps, we propose a matched swap pair construction method to bound the approximation loss, ensuring a constant probability of loss reduction in each local search step. In expectation, these techniques enable us to obtain the local search algorithm for the CSS problem with theoretical guarantees, where a 53(k+1)-approximate solution can be obtained in linear running time O(ndk^4\log k). Empirical experiments show that our proposed algorithm achieves better quality and time compared to previous algorithms on both small and large datasets. Moreover, it is at least 10 times faster than state-of-the-art algorithms across all large-scale datasets.

Protein-ligand interactions (PLIs) are essential for cellular activities and drug discovery. But due to the complexity and high cost of experimental methods, there is a great demand for computational approaches to recognize PLI patterns, such as protein-ligand docking. In recent years, more and more models based on machine learning have been developed to directly predict the root mean square deviation (RMSD) of a ligand docking pose with reference to its native binding pose. However, new scoring methods are pressingly needed in methodology for more accurate RMSD prediction. We present a new deep learning-based scoring method for RMSD prediction of protein-ligand docking poses based on a Graphormer method and Shell-like graph architecture, named GSScore. To recognize near-native conformations from a set of poses, GSScore takes atoms as nodes and then establishes the docking interface of protein-ligand into multiple bipartite graphs within different shell ranges. Benefiting from the Graphormer and Shell-like graph architecture, GSScore can effectively capture the subtle differences between energetically favorable near-native conformations and unfavorable non-native poses without extra information. GSScore was extensively evaluated on diverse test sets including a subset of PDBBind version 2019, CASF2016 as well as DUD-E, and obtained significant improvements over existing methods in terms of RMSE, $R$ (Pearson correlation coefficient), Spearman correlation coefficient and Docking power.

Human leukocyte antigen (HLA) recognizes foreign threats and triggers immune responses by presenting peptides to T cells. Computationally modeling the binding patterns between peptide and HLA is very important for the development of tumor vaccines. However, it is still a big challenge to accurately predict HLA molecules binding peptides. In this paper, we develop a new model TripHLApan for predicting HLA molecules binding peptides by integrating triple coding matrix, BiGRU + Attention models, and transfer learning strategy. We have found the main interaction site regions between HLA molecules and peptides, as well as the correlation between HLA encoding and binding motifs. Based on the discovery, we make the preprocessing and coding closer to the natural biological process. Besides, due to the input being based on multiple types of features and the attention module focused on the BiGRU hidden layer, TripHLApan has learned more sequence level binding information. The application of transfer learning strategies ensures the accuracy of prediction results under special lengths (peptides in length 8) and model scalability with the data explosion. Compared with the current optimal models, TripHLApan exhibits strong predictive performance in various prediction environments with different positive and negative sample ratios. In addition, we validate the superiority and scalability of TripHLApan's predictive performance using additional latest data sets, ablation experiments and binding reconstitution ability in the samples of a melanoma patient. The results show that TripHLApan is a powerful tool for predicting the binding of HLA-I and HLA-II molecular peptides for the synthesis of tumor vaccines. TripHLApan is publicly available at https://github.com/CSUBioGroup/TripHLApan.git.

MOTIVATION: Identifying cancer genes remains a significant challenge in cancer genomics research. Annotated gene sets encode functional associations among multiple genes, and cancer genes have been shown to cluster in hallmark signaling pathways and biological processes. The knowledge of annotated gene sets is critical for discovering cancer genes but remains to be fully exploited. RESULTS: Here, we present the DIsease-Specific Hypergraph neural network (DISHyper), a hypergraph-based computational method that integrates the knowledge from multiple types of annotated gene sets to predict cancer genes. First, our benchmark results demonstrate that DISHyper outperforms the existing state-of-the-art methods and highlight the advantages of employing hypergraphs for representing annotated gene sets. Second, we validate the accuracy of DISHyper-predicted cancer genes using functional validation results and multiple independent functional genomics data. Third, our model predicts 44 novel cancer genes, and subsequent analysis shows their significant associations with multiple types of cancers. Overall, our study provides a new perspective for discovering cancer genes and reveals previously undiscovered cancer genes. AVAILABILITY AND IMPLEMENTATION: DISHyper is freely available for download at https://github.com/genemine/DISHyper.

MOTIVATION: Studying the molecular heterogeneity of cancer is essential for achieving personalized therapy. At the same time, understanding the biological processes that drive cancer development can lead to the identification of valuable therapeutic targets. Therefore, achieving accurate and interpretable clinical predictions requires paramount attention to thoroughly characterizing patients at both the molecular and biological pathway levels. RESULTS: Here, we present GraphPath, a biological knowledge-driven graph neural network with multi-head self-attention mechanism that implements the pathway-pathway interaction network. We train GraphPath to classify the cancer status of patients with prostate cancer based on their multi-omics profiling. Experiment results show that our method outperforms P-NET and other baseline methods. Besides, two external cohorts are used to validate that the model can be generalized to unseen samples with adequate predictive performance. We reduce the dimensionality of latent pathway embeddings and visualize corresponding classes to further demonstrate the optimal performance of the model. Additionally, since GraphPath's predictions are interpretable, we identify target cancer-associated pathways that significantly contribute to the model's predictions. Such a robust and interpretable model has the potential to greatly enhance our understanding of cancer's biological mechanisms and accelerate the development of targeted therapies. AVAILABILITY AND IMPLEMENTATION: https://github.com/amazingma/GraphPath.

MOTIVATION: Nanopore direct RNA sequencing (DRS) enables the detection of RNA N6-methyladenosine (m6A) without extra laboratory techniques. A number of supervised or comparative approaches have been developed to identify m6A from Nanopore DRS reads. However, existing methods typically utilize either statistical features of the current signals or basecalling-error features, ignoring the richer information of the raw signals of DRS reads. RESULTS: Here, we propose RedNano, a deep-learning method designed to detect m6A from Nanopore DRS reads by utilizing both raw signals and basecalling errors. RedNano processes the raw-signal feature and basecalling-error feature through residual networks. We validated the effectiveness of RedNano using synthesized, Arabidopsis, and human DRS data. The results demonstrate that RedNano surpasses existing methods by achieving higher area under the ROC curve (AUC) and area under the precision-recall curve (AUPRs) in all three datasets. Furthermore, RedNano performs better in cross-species validation, demonstrating its robustness. Additionally, when detecting m6A from an independent dataset of Populus trichocarpa, RedNano achieves the highest AUC and AUPR, which are 3.8%-9.9% and 5.5%-13.8% higher than other methods, respectively. AVAILABILITY AND IMPLEMENTATION: The source code of RedNano is freely available at https://github.com/Derryxu/RedNano.

MOTIVATION: Mutations are the crucial driving force for biological evolution as they can disrupt protein stability and protein-protein interactions which have notable impacts on protein structure, function, and expression. However, existing computational methods for protein mutation effects prediction are generally limited to single point mutations with global dependencies, and do not systematically take into account the local and global synergistic epistasis inherent in multiple point mutations. RESULTS: To this end, we propose a novel spatial and sequential message passing neural network, named DDAffinity, to predict the changes in binding affinity caused by multiple point mutations based on protein 3D structures. Specifically, instead of being on the whole protein, we perform message passing on the k-nearest neighbor residue graphs to extract pocket features of the protein 3D structures. Furthermore, to learn global topological features, a two-step additive Gaussian noising strategy during training is applied to blur out local details of protein geometry. We evaluate DDAffinity on benchmark datasets and external validation datasets. Overall, the predictive performance of DDAffinity is significantly improved compared with state-of-the-art baselines on multiple point mutations, including end-to-end and pre-training based methods. The ablation studies indicate the reasonable design of all components of DDAffinity. In addition, applications in nonredundant blind testing, predicting mutation effects of SARS-CoV-2 RBD variants, and optimizing human antibody against SARS-CoV-2 illustrate the effectiveness of DDAffinity. AVAILABILITY AND IMPLEMENTATION: DDAffinity is available at https://github.com/ak422/DDAffinity.

MOTIVATION: Haplotype-resolved genome assemblies serve as vital resources in various research domains, including genomics, medicine, and pangenomics. Algorithms employing Hi-C data to generate haplotype-resolved assemblies are particularly advantageous due to its ready availability. Existing methods primarily depend on mapping quality to filter out uninformative Hi-C alignments which may be susceptible to sequencing errors. Setting a high mapping quality threshold filters out numerous informative Hi-C alignments, whereas a low mapping quality threshold compromises the accuracy of Hi-C alignments. Maintaining high accuracy while retaining a maximum number of Hi-C alignments can be challenging. RESULTS: In our experiments, heterozygous variations play an important role in filtering uninformative Hi-C alignments. Here, we introduce Diphase, a novel phasing tool that harnesses heterozygous variations to accurately identify the informative Hi-C alignments for phasing and to extend primary/alternate assemblies. Diphase leverages mapping quality and heterozygous variations to filter uninformative Hi-C alignments, thereby enhancing the accuracy of phasing and the detection of switches. To validate its performance, we conducted a comparative analysis of Diphase, FALCON-Phase, and GFAse on various human datasets. The results demonstrate that Diphase achieves a longer phased block N50 and exhibits higher phasing accuracy while maintaining a lower hamming error rate. AVAILABILITY AND IMPLEMENTATION: The source code of Diphase is available at https://github.com/zhangjuncsu/Diphase.

MOTIVATION: Dropout events bring challenges in analyzing single-cell RNA sequencing data as they introduce noise and distort the true distributions of gene expression profiles. Recent studies focus on estimating dropout probability and imputing dropout events by leveraging information from similar cells or genes. However, the number of dropout events differs in different cells, due to the complex factors, such as different sequencing protocols, cell types, and batch effects. The dropout event differences are not fully considered in assessing the similarities between cells and genes, which compromises the reliability of downstream analysis. RESULTS: This work proposes a hybrid Generative Adversarial Network with dropouts identification to impute single-cell RNA sequencing data, named AGImpute. First, the numbers of dropout events in different cells in scRNA-seq data are differentially estimated by using a dynamic threshold estimation strategy. Next, the identified dropout events are imputed by a hybrid deep learning model, combining Autoencoder with a Generative Adversarial Network. To validate the efficiency of the AGImpute, it is compared with seven state-of-the-art dropout imputation methods on two simulated datasets and seven real single-cell RNA sequencing datasets. The results show that AGImpute imputes the least number of dropout events than other methods. Moreover, AGImpute enhances the performance of downstream analysis, including clustering performance, identifying cell-specific marker genes, and inferring trajectory in the time-course dataset. AVAILABILITY AND IMPLEMENTATION: The source code can be obtained from https://github.com/xszhu-lab/AGImpute.

Abstract. In this paper, we study the following Minimum Cost Multicovering (MCMC) problem: Given a set of [Formula: see text] client points [Formula: see text] and a set of [Formula: see text] server points [Formula: see text] in a fixed dimensional [Formula: see text] space, determine a set of disks centered at these server points so that each client point [Formula: see text] is covered by at least [Formula: see text] disks and the total cost of these disks is minimized, where [Formula: see text] is a function that maps every client point to some nonnegative integer no more than [Formula: see text] and the cost of each disk is measured by the [Formula: see text]th power of its radius for some constant [Formula: see text]. MCMC is a fundamental optimization problem with applications in many areas such as wireless/sensor networking. Despite extensive research on this problem for about two decades, only constant approximations were known for general [Formula: see text]. It has been a long standing open problem to determine whether a PTAS is possible. In this paper, we give an affirmative answer to this question by presenting the first PTAS for it. Our approach is based on a number of novel techniques, such as balanced recursive realization and bubble charging, and new counterintuitive insights to the problem. Particularly, we approximate each disk with a set of sub-boxes and optimize them at the subdisk level. This allows us to first compute an approximate disk cover through dynamic programming, and then obtain the desired disk cover through a balanced recursive realization procedure.

Handwriting images are commonly used to diagnose Parkinson's disease due to their intuitive nature and easy accessibility. However, existing methods have not explored the potential of the fusion of different handwriting image sources for diagnosis. To address this issue, this study proposes a hybrid fusion approach that makes use of the visual information derived from different handwriting images and handwriting templates, significantly enhancing the performance in diagnosing Parkinson's disease. The proposed method involves several key steps. Initially, different preprocessed handwriting images undergo pixel-level fusion using Laplacian transformation. Subsequently, the fused and original images are fed into a pre-trained CNN separately to extract visual features. Finally, feature-level fusion is performed by concatenating the feature vectors extracted from the flatten layer, and the fused feature vectors are input into SVM to obtain classification results. Our experimental results validate that the proposed method achieves excellent performance by only utilizing visual features from images, with 95.45% accuracy on the NewHandPD. Furthermore, the results obtained on our dataset verify the strong generalizability of the proposed approach.

Assigning appropriate rhetorical roles, such as "background," "intervention," and "outcome," to sentences in biomedical documents can streamline the process for physicians to locate evidence and resources for medical treatment and decision-making. While sequence labeling and span-based methods are frequently employed for this task, the former disregards a document's semantic structure, resulting in a lack of semantic coherence across continuous sentences. Span-based approaches, on the other hand, either necessitate the enumeration of all potential spans, which can be time-consuming, or may lead to the misclassification of sentences over extended spans. Consequently, an approach is required that models the semantic structure of documents explicitly and captures boundary information to achieve precise and effective sentence labeling in biomedical documents. To address these challenges, we propose a new approach, the boundary-aware dual biaffine model, which explicitly models the semantic structure of documents and incorporates boundary information via a dual biaffine layer. We introduce a dynamic programming algorithm to minimize missing labels and overlapping predictions, and achieve globally optimal decoding results. We evaluate our approach on three benchmark datasets, namely PubMed 20 k RCT, PubMed-PICO and NICTA-PIBOSO. The experimental results demonstrate that our approach outperforms strong baselines and achieves state-of-the-art performance on PubMed 20 k RCT and PubMed-PICO. Additionally, our method also achieves competitive results on NICTA-PIBOSO.

Chinese electronic medical record (EMR) presents significant challenges for named entity recognition (NER) due to their specialized nature, unique language features, and diverse expressions. Traditionally, NER is treated as a sequence labeling task, where each token is assigned a label. Recent research has reframed NER within the machine reading comprehension (MRC) framework, extracting entities in a question-answer format, achieving state-of-the-art performance. However, these MRC-based methods have a significant limitation: they extract entities of various types independently, ignoring their interrelations. To address this, we introduce the Fusion Label Relations with MRC (FLR-MRC) model, which enhances the MRC model by implicitly capturing dependencies among entity types. FLR-MRC models interrelations between labels using graph attention networks, integrating these with textual data to identify entities. On the benchmark CMeEE and CCKS2017-CNER datasets, FLR-MRC achieves F1-scores of 0.6652 and 0.9101, respectively, outperforming existing clinical NER methods.

Piwi-interacting RNAs (piRNAs) are increasingly recognized as potential biomarkers for various diseases. Investig-ating the complex relationship between piRNAs and diseases through computational methods can reduce the costs and risks associated with biological experiments. Fast kernel learning (FKL) is a classical method for multi-source data fusion that is widely employed in association prediction research. However, biological networks are noisy due to the limitations of measurement technology and inherent natural variation, which can hamper the effectiveness of the network-based ideal kernel. The conventional FKL method does not address this issue. In this study, we propose a low-rank fast kernel learning (LRFKL) algorithm, which consists of low-rank representation (LRR) and the FKL algorithm. The LRFKL algorithm is designed to mitigate the effects of noise on the network-based ideal kernel. Using LRFKL, we propose a novel approach for predicting piRNA-disease associations called LKLPDA. Specifically, we first compute the similarity matrices for piRNAs and diseases. Then we use the LRFKL to fuse the similarity matrices for piRNAs and diseases separately. Finally, the LKLPDA employs AutoGluon-Tabular for predictive analysis. Computational results show that LKLPDA effectively predicts piRNA-disease associations with higher accuracy compared to previous methods. In addition, case studies confirm the reliability of the model in predicting piRNA-disease associations.

Precisely predicting Drug-Drug Interactions (DDIs) carries the potential to elevate the quality and safety of drug therapies, protecting the well-being of patients, and providing essential guidance and decision support at every stage of the drug development process. In recent years, leveraging large-scale biomedical knowledge graphs has improved DDI prediction performance. However, the feature extraction procedures in these methods are still rough. More refined features may further improve the quality of predictions. To overcome these limitations, we develop a knowledge graph-based method for multi-typed DDI prediction with contrastive learning (KG-CLDDI). In KG-CLDDI, we combine drug knowledge aggregation features from the knowledge graph with drug topological aggregation features from the DDI graph. Additionally, we build a contrastive learning module that uses horizontal reversal and dropout operations to produce high-quality embeddings for drug-drug pairs. The comparison results indicate that KG-CLDDI is superior to state-of-the-art models in both the transductive and inductive settings. Notably, for the inductive setting, KG-CLDDI outperforms the previous best method by 17.49% and 24.97% in terms of AUC and AUPR, respectively. Furthermore, we conduct the ablation analysis and case study to show the effectiveness of KG-CLDDI. These findings illustrate the potential significance of KG-CLDDI in advancing DDI research and its clinical applications.

Accurate identification of protein-protein interaction (PPI) sites is crucial for understanding the mechanisms of biological processes, developing PPI networks, and detecting protein functions. Currently, most computational methods primarily concentrate on sequence context features and rarely consider the spatial neighborhood features. To address this limitation, we propose a novel residual graph convolutional network for structure-based PPI site prediction (RGCNPPIS). Specifically, we use a GCN module to extract the global structural features from all spatial neighborhoods, and utilize the GraphSage module to extract local structural features from local spatial neighborhoods. To the best of our knowledge, this is the first work utilizing local structural features for PPI site prediction. We also propose an enhanced residual graph connection to combine the initial node representation, local structural features, and the previous GCN layer's node representation, which enables information transfer between layers and alleviates the over-smoothing problem. Evaluation results demonstrate that RGCNPPIS outperforms state-of-the-art methods on three independent test sets. In addition, the results of ablation experiments and case studies confirm that RGCNPPIS is an effective tool for PPI site prediction.

Multi-path Transmission Control Protocol (MPTCP) has shown great potential in improving network bandwidth and robustness by utilizing multiple subflows in data center networks (DCNs). However, the delay and loss heterogeneities of multiple paths potentially cause packet reordering, resulting in the ACK blocking and increased latency. Recent coding-based solutions use forward error correction (FEC) to mitigate path heterogeneity with redundant encoded packets. However, current FEC-based solutions work at the subflow level, that is, each subflow independently generates redundant encoded packets. This intra-subflow coding, however, does not leverage the path diversity, easily suffering from long tail latency. In this article, we propose a new MPTCP based on cyclic matrix coding, called as CM-MPTCP, which encodes packets inter subflow to leverage the path diversity. Specifically, to let good paths help bad ones, the good paths deliver more redundant packets encoded based on cyclic matrix, which gives more coding opportunities to packets on bad paths, thus achieving high packet decoding ratio at the receiver side. The results of large-scale NS2 simulations show that CM-MPTCP effectively mitigates the ACK blocking and reduces the average flow completion time (AFCT) by about$45\% \sim 70\%$under a wide variety of network conditions compared with the state-of-the-art coding-based MPTCPs.

Remote Direct Memory Access (RDMA) achieves ultra-low latency, high throughput and low CPU overhead in data center by implementing the transport logic in hardware network interface card (NIC). However, RDMA faces new challenges in the heterogeneous multipath environment as it is very sensitive to packet reordering. When some packets are blocked in slow paths, the other packets delivered through fast paths have to be buffered at the receiver's NIC, consuming the limited on-chip memory resources. In this paper, we propose a new RDMA-based multipath transmission scheme with advanced fast retransmission called as AFR-MPRDMA. Specifically, once detecting congestion at the slow path, the sender will retransmit the blocked packets on other fast paths to speed up the transmission of blocked packets. Moreover, the receiver dynamically adjusts the buffer size for the out-of-order packets to avoid either unnecessary retransmission or long latency. The results of large-scale tests show that AFR-MPRDMA effectively mitigates packets blocking issue and reduces average flow completion time (AFCT) by up to 61% compared with the state-of-the-art RDMA-based schemes.

The diagnosis of mild cognitive impairment (MCI), which is an early stage of Alzheimer’s disease (AD), has great clinical significance. Medical imaging and gene sequencing technologies have provided sufficient multimodality data for MCI diagnostic studies. However, how to effectively extract the rich representations from multimodality data remains a challenging task. To address this challenging task, we propose a new multimodality multiview graph representations and knowledge embedding (MMGK) framework to diagnose MCI. First, to obtain rich information from multimodality data, we extract multiview feature representations from magnetic resonance imaging (MRI) and genetic data. Afterward, considering the correlations between subjects, all subjects are constructed into a graph based on the different single-view feature representations, respectively. To further enrich the correlations between subjects, demographic data are utilized through knowledge embedding. Finally, to perform MCI diagnosis on multiview graphs, graph convolutional networks (GCNs) are utilized. In addition, to further improve the performance of MCI diagnosis, a two-step ensemble learning method is proposed. The proposed framework is evaluated on 188 subjects from the AD Neuroimaging Initiative (ADNI). Experimental results show that our proposed framework achieves good performance with accuracy reaching 0.888, and outperforms some state-of-the-art (SOTA) methods. In addition, the proposed framework is applied to Parkinson’s disease (PD) diagnosis and achieves 0.856 accuracy. Overall, our proposed method has potential for clinical application in MCI diagnosis and other diseases via integrating MRI, genetic data, and demographic data. Our code is available at:https://github.com/miacsu/MMGK.

The Object goal Navigation (ObjectNav) task requires an agent to navigate through a previously unknown domestic scenario using spatial and semantic contextual information, where the goal is specified by a semantic label (e.g., find a TV). Such a task is especially challenging as it requires formulating and understanding the complex co-occurrence relations among objects in diverse settings, which is critical for long-sequence navigational decision-making. Existing methods learn to either explicitly represent co-occurrence relationships as discrete semantic priors, or implicitly encode them from raw observations, thus can not benefit from the rich environmental semantics. In this work, we propose a novel Deep Reinforcement Learning (DRL) based ObjectNav strategy by actively imagining spatial and semantic clues outside the agent’s Field of View (FoV) and further mining Continuous Environmental Representations (CER) using self-supervised learning. Additionally, the illusion of spatial and semantic patterns allows the agent to perform Multi-Step Forward-Looking Planning (MSFLP) by considering the temporal evolution of egocentric local observations. Our approach is thoroughly evaluated and ablated in the visually realistic environments of the Matterport3D (MP3D) dataset. The experimental results reflect that our method combining CER and imagination-based MSFLP facilitates learning complicated semantic priors and navigation skills, thus achieving state-of-the-art performance on the ObjectNav task. In addition, adequate quantitative and qualitative analyses validate the excellent generalization ability and superiority of our method.

Mobile video streaming dominates today's network traffic, and adaptive bitrate (ABR) algorithms have been routinely adopted for transmitting media content across dynamic mobile networks. State-of-the-art ABR algorithms mainly alter video bitrate without considering audio bitrate as they consider the impact on the video negligible due to their small size. However, to bring users an immersive experience, recent content providers have applied high-quality audio with large sizes, like stereophonic sound. Therefore, improper audio bitrate selection will adversely affect video bitrate selection, leading to undesirable audio/video combinations (the highest video quality with the lowest audio quality, and vice versa) and frequent playback interruptions. To address these inefficiencies, we propose a Self-Play reinforcement learning-based Audio-aware ABR algorithm named SPA to learn strategies for audio and video bitrate selections. By learning from explicit goals, SPA can match the actual requirements and attain good performance. By conducting trace-driven and testbed-based experiments, we observe SPA's considerable superiority compared to existing approaches, including reducing the undesirable combinations by up to 34.17× and achieving zero stall time across 88.57% of traces. We also invite 35 volunteers to join a subjective test, and the result shows that 33/35 people consider SPA provides them with a satisfactory viewing experience.

Adaptive Bitrate (ABR) algorithms have become increasingly important for delivering high-quality video content over fluctuating networks. Considering the complexity of video scenes, video chunks can be separated into two categories: those with intricate scenes and those with simple scenes. In practice, it has been observed that improving the quality of intricate chunks yields more substantial improvements in Quality of Experience (QoE) compared with focusing solely on simple chunks. However, the current ABR schemes either treat all chunks equally or rely on fixed linear-based reward functions, which limits their ability to meet real-world requirements. To tackle these limitations, this paper introduces a novel ABR approach called CAST (Complex-scene Aware bitrate algorithm via Self-play reinforcemenT learning), which considers the scene complexity and formulates the bitrate adaptation task as an explicit objective. Leveraging the power of parallel computing with multiple agents, CAST trains a neural network to achieve superior video playback quality for intricate scenes while minimizing playback freezing time. Moreover, we also introduce a new variant of our proposed approach called CAST-DU, to address the critical issue of efficiently managing users' limited cellular data budgets while ensuring a satisfactory viewing experience. Furthermore, we present CAST-Live, tailored for live streaming scenarios with constrained playback buffers and considerations for energy costs. Extensive trace-driven evaluations and subjective tests demonstrate that CAST, CAST-DU, and CAST-Live outperform existing off-the-shelf schemes, delivering a superior video streaming experience over fluctuating networks while efficiently utilizing data resources. Moreover, CAST-Live demonstrates effectiveness even under limited buffer size constraints while incurring minimal energy costs.

Adaptive BitRate (ABR) algorithms have become increasingly prevalent in modern streaming platforms, offering users significant improvements in the Quality of Experience (QoE). With streaming providers like YouTube and Netflix shifting to high-fidelity audio formats such as stereophonic sound and Dolby Atoms, ensuring proper audio and video adaptation has become a critical aspect of modern streaming platforms. Additionally, Variable Bitrate (VBR) encoding has gained great popularity in encoding audio and video content, given its higher quality-to-bits ratio. However, the considerable variability in network bandwidth, in combination with VBR features such as significantly fluctuating audio/video chunk sizes and diverse content complexity, makes existing ABR schemes formidable to make optimal bitrate selection due to their overlook of audio adaptation or oblivious to VBR features. In this paper, we introduce a new ABR approach forVBR-basedAudio-aware videoStrEaming named VASE, which harnesses deep reinforcement learning (DRL) and exploits parallel computing with multiple agents to swiftly and adeptly manage fluctuations in video/audio chunk sizes, network bandwidth, and varying content complexity, all while operating without any assumptions. Besides, two variants are proposed to mitigate the download energy cost and handle audio and video content in finer granularity. Extensive trace-driven, testbed, and subjective evaluations show that our scheme surpasses existing advanced adaptation schemes regarding the overall QoE, effectively demonstrating its superiority.

Performing accurate sensing in diverse environments is a challenging issue in wireless sensing technologies. Existing solutions usually require collecting a large number of samples to train a classifier for every environment, or further assume similar sample distribution between different environments such that a model trained in one environment can be transferred to another. In this paper, we propose RF-Siamese, an RFID-based gesture sensing approach that achieves comparable accuracy to existing solutions but requires only a few samples in each eivironment. RF-Siamese leverages Siamese networks to distinguish different gestures with only a small number of samples and is enhanced by several novel designs to achieve high accuracy in diverse environments. First, the network structure and parameters (e.g., loss function and distance metric) are carefully designed to be suitable for RFID gesture recognition. Second, a permutation-based dataset generation strategy is proposed to make full use of the collected samples to enhance the recognition accuracy. Third, a template matching method is proposed to extend the Siamese network to classify multiple gestures. Extensive experiments on commercial RFID devices demonstrate that RF-Siamese achieves a high accuracy of 0.93 with only one sample of each gesture when recognizing 18 different gestures, while state-of-the-art approaches based on transfer learning and meta learning achieve an accuracy of only 0.59 and 0.70, respectively.

Lesion segmentation is a fundamental step for the diagnosis of acute ischemic stroke (AIS). Non-contrast CT (NCCT) is still a mainstream imaging modality for AIS lesion measurement. However, AIS lesion segmentation on NCCT is challenging due to low contrast, noise and artifacts. To achieve accurate AIS lesion segmentation on NCCT, this study proposes a hybrid convolutional neural network (CNN) and Transformer network with circular feature interaction and bilateral difference learning. It consists of parallel CNN and Transformer encoders, a circular feature interaction module, and a shared CNN decoder with a bilateral difference learning module. A new Transformer block is particularly designed to solve the weak inductive bias problem of the traditional Transformer. To effectively combine features from CNN and Transformer encoders, we first design a multi-level feature aggregation module to combine multi-scale features in each encoder and then propose a novel feature interaction module containing circular CNN-to-Transformer and Transformer-to-CNN interaction blocks. Besides, a bilateral difference learning module is proposed at the bottom level of the decoder to learn the different information between the ischemic and contralateral sides of the brain. The proposed method is evaluated on three AIS datasets: the public AISD, a private dataset and an external dataset. Experimental results show that the proposed method achieves Dices of 61.39% and 46.74% on the AISD and the private dataset, respectively, outperforming 17 state-of-the-art segmentation methods. Besides, volumetric analysis on segmented lesions and external validation results imply that the proposed method is potential to provide support information for AIS diagnosis.

In tile-based 360° video streaming, the users employ the tile rate allocation algorithm to select appropriate bitrate to maximize the quality of experience (QoE). The preferences and viewports, however, can vary significantly across the different users. Since the users independently choose their bitrate according to their own preferences and viewports, it is hard to ensure QoE fairness for users under the constraint of available bandwidth. In this article, we propose a QoE-fairness aware bitrate allocation algorithm for multi-users (QBAM) to reduce difference of user QoE. According to the trajectory of the user viewpoint and user preferences for video quality, rebuffer time and quality switching, we leverage multi-agent reinforcement learning to train the bitrate allocation strategy. The experimental results show, compared with the current tile rate allocation algorithm, QBAM effectively improves the QoE fairness.

Delay-based congestion control has drawn a lot of attention in both academics and industry recently. Specifically, the Copa algorithm proposed in NSDI can achieve consistent high performance under various network environments and has already been deployed on Facebook. In this paper, we theoretically analyze Copa and reveal its large queuing delay and poor fairness issue under certain conditions. The root cause is that Copa fails to achieve its expected behaviors, i.e., clear the bottleneck buffer occupancy periodically. Moreover, we also reveal that the pathological competitive mode of Copa fails to guarantee friendliness. To address these issues, we propose Copa+, which enhances Copa with a parameter adaptation mechanism and an optimized competitive mode. Designed based on our theoretical analysis, Copa+ can adaptively clear the bottleneck buffer occupancy and become friendly to Cubic in the competitive mode. As a result, Copa+ inherits the advantages of Copa but achieves lower queuing delay and better fairness under different environments, as confirmed by real-world experiments and simulations. Specifically, Copa+ has the highest average throughput over different Internet links among different cloud nodes, compared to Cubic, BBR, PCC Vivace, Remy, and Indigo. Meanwhile, Copa+ has an 8.1% increase in throughput and similar low queuing delay compared to Copa. Moreover, Copa+ achieves 14.6% lower queuing delay and 2.4% higher throughput compared to Sprout over emulated cellular links.

Deep Neural Network (DNN) is a critical component of a wide range of applications. However, with the rapid growth of the training dataset and model size, communication becomes the bottleneck, resulting in low utilization of computing resources. To accelerate communication, recent works propose to aggregate gradients from multiple workers in the programmable switch to reduce the volume of exchanged data. Unfortunately, since using synchronization transmission to aggregate data, current in-network aggregation designs suffer from the straggler problem, which often occurs in shared clusters due to resource contention. To address this issue, we propose a straggler-aware aggregation transport protocol (SA-ATP), which enables the leading worker to leverage the spare computing and storage resources to help the straggling worker. We implement SA-ATP atop clusters using P4-programmable switches. The evaluation results show that SA-ATP reduces the iteration time by up to 57% and accelerates training by up to$1.8\times $in real-world benchmark models.

To achieve low latency and high link utilization, ECN-based transport protocols (i.e., DCTCP) are widely deployed in data center networks (DCN). In multi-tenant environment, however, the newly introduced ECN-enabled TCP greatly impairs the performance of applications with out-dated and misconfigured TCP stacks. The reason is that the ECN-enabled switch fails to treat the mixed TCP traffic fairly, resulting in the distinguished performance gap between the ECN-enabled and ECN-disabled TCPs. This paper proposes DDT (Dual Dynamic Thresholds), an active queue management algorithm (AQM) to achieve the flow-level fairness for coexisting heterogeneous TCP traffic. DDT monitors the switch queue in real time, and dynamically tunes the distance between ECN-marking and packet-dropping thresholds to mitigate the aggressiveness difference between the ECN-enabled and ECN-disabled TCPs. The results of real implementations and large-scaled simulations show that DDT elegantly fills the aggressiveness gap of heterogeneous TCP traffic without disturbing their own control loops, while only introducing acceptable deployment overhead at switch.

During outbreaks of large infectious diseases like COVID-19, there is a strain on healthcare resources worldwide. To alleviate the burden on healthcare workers during the initial stages of the outbreak, there is an urgent need for the development of automated tools. Federated learning offers a privacypreserving solution to the challenge that limited annotated data within single healthcare facilities. However, existing federated aggregation strategies cannot adapt to the real-world medical image data problem of mixed heterogeneous. This paper introduces Personalized Federated Adaptive Regularization (pFedAR), an adaptive framework for federated learning that effectively utilizes multi-site COVID-19 CT datasets with mixed heterogeneous of distribution discrepancies. To solve the mixed distribution of label skew and feature shift, we propose a two-stage adaptive regularization. In the client training stage, we use the balance loss term to balance the COVID-19 clients with missing label. In the aggregation stage, we correct the client gradient conflict. Our method is developed and evaluated using eight real-world COVID-19 diagnosis datasets composed of CT images. Extensive experiments demonstrate the consistent improvement achieved by our method across the datasets.

Medical image denoising, as a part of medical image processing, is significant for the assessment and diagnosis of diseases. To improve the medical image denoising performance of existing deep learning methods, we propose a lightweight plug-and-play module (LP2M) with low complexity, which can be plugged before current image denoising methods. Specifically, the proposed LP2M contains three stacked Convolutional Neural Network (CNN) based blocks: a image receptor block, an adaptive receptive field selection block and a high-low frequency processing block. The image receptor block can perceive color or grayscale images and perform preliminary processing. The adaptive receptive field selection block includes two parallel paths with different receptive fields (i.e., convolution kernel sizes) and the adaptive weighting operation, which can process multi-scale information in the image. The high-low frequency processing block consists of a low frequency pathway using convolutional layers with large kernel sizes, and a high frequency pathway using covolutional layers with small kernel sizes, which can process the low and high frequency components in images. Extensive validation experiments are performed on five state-of-the-art denoising methods on multiple medical image datasets for three different medical image denoising tasks (X-ray image denoising, magnetic resonance image denoising and dermoscopic image denoising). Experimental results show that our proposed LP2M can effectively improve the results of these five state-of-the-art methods for three denoising tasks with only increasing 0.996K parameters and 63.112M FLOPs, and it can provide potential direction for improving image denoising.

Self-supervised learning provides a possible solution to extract effective visual representations from unlabeled pathological images. However, most of the existing methods either do not effectively utilize domain-specific information or are designed and optimized for image classification, resulting in these pre-trained models that may not be optimal for pathological image segmentation. In this paper, we propose DKSL: Domain-specific Knowledge guided Self-supervised Learning, which uses image reconstruction tasks to aid contrastive learning and exploits single-dye stained pathological images after stain separation as domain-specific knowledge to guide the model. Our method provides a novel way to exploit the domain-specific knowledge of pathological images. In contrastive learning, we add single-dye stained images as an expansion of the original positive samples to the contrastive learning process to preserve more global semantic information. In image reconstruction, the model is forced to focus on local image details relevant to downstream tasks by reconstructing single-dye stained images from the representation extracted by the encoder of contrastive learning. Finally, the encoder and decoder from the pre-training stage are fine-tuned by the downstream segmentation task. Fine-tuning experimental results demonstrate that DKSL outperforms state-of-the-art methods with Dices of 90.50% and 79.68% on two publicly available datasets, GlaS and MoNuSeg, respectively.

Isocitrate dehydrogenase (IDH) is a key molecular feature for gliomas, and the prediction of IDH is also an important task for computer-aided diagnosis using magnetic resonance imaging (MRI). To address this changllenge, we introduce a multi-modal MRI-based characteristics inspired network for IDH Genotyping (M3CI-Net), which pay more attention to the different characteristics information of different MRI modalities T1, T2, T1ce, Flair. In M3CI-Net, a pre-fusion module with multi-channel attention mechanism is used to fuse T1ce and Flair modalities and capture as much as possible luminance and contrast information, and the edge information is obtained from T2 modality by using edge detection module. Finally, the feature information between modalities are fused and input into a CNN-Transformer based encoder structure to extract shared spatial and global information from multi-modal MRI, and the information of multiple scales frome encoder are input into the linear layer for IDH genotype classification after pooling, meanwhile, the CNN based decoder with skip-connection for glioma segmentation works for assisting IDH genotyping. Then, we proposed images’ pre-fusion loss, segmentation loss, IDH genotyping loss, and use uncertainty weight training method to balance the weights of these loss. we evaluate our proposed method on Brats2020, and achieve an acceracy of 0.88, an AUC of 0.94, a specificity of 0.92, a sensitivity of 0.84 in IDH genotyping, which is superior to the state-of-the-art methods.

Accurately predicting survival of esophageal cancer is essential for clinical precision treatment. However, the existing region of interest (ROI) based methods not only require prior medical knowledge to complete the delineation of tumor, but may also lead to excessive sensitivity of the model towards ROI. To address these challenges, we design a fully automated CT-guided learning that combines a CNN-Transformer size aware U-Net and a ranked survival prediction network together to automatically predict the survival of patients with esophageal cancer. Specifically, we first incorporate the Transformer with shifted windowing multi-head self-attention mechanism into the base of the encoder in the U-Net to capture the long-range dependency in the 3D CT images. Then, to alleviate the imbalance between the ROI and the background in CT images, we design a size-aware coefficient for the segmentation loss. Finally, we design a ranked pair sorting loss to learn more fully the ranked information hidden in esophageal cancer patients. To validate the effectiveness of our method, we conduct extensive experiments on a dataset containing 759 esophageal cancer samples. The experimental results demonstrate that our proposed method can still achieve the best performance in survival prediction without ROI ground truth.

Inferring the fine-grained urban flows based on the coarse-grained flow observations is practically important to many smart city-related applications. However, the collected human/vehicle trajectory flows are usually rather unreliable, may contain various noise and sometimes are incomplete, thus posing great challenges to existing approaches. In this paper, we present a pioneering study on robust fine-grained urban flow inference with noisy and incomplete urban flow observations, and propose a denoising diffusion model named DiffUFlow to effectively address it. Specifically, we propose an improved reverse diffusion strategy. A spatial-temporal feature extraction network called STFormer and a semantic features extraction network called ELFetcher are also proposed. Then, we overlay the spatial-temporal feature map extracted by STFormer onto the coarse-grained flow map, serving as a conditional guidance for the reverse diffusion process. We further integrate the semantic features extracted by ELFetcher to cross-attention layers, enabling the comprehensive consideration of semantic information encompassing the entirety of urban data in fine-grained inference. Extensive experiments on two large real-world datasets validate the effectiveness of our method compared with the state-of-the-art baselines.

Semantic text matching has a wide range of applications in natural language processing. Recently proposed models that have achieved excellent results on short text matching tasks are not well suited to long-form text matching problems due to input length limitations and increased noise. On the other hand, long-form texts contain a large amount of information at different granularities after encoding, which cannot be fully interacted and utilized by existing methods. To address above issues, we propose a novel long-form text-matching framework which fuses Bi-Encoder and Cross-Encoder (FBC). Specially, it first employs an entity-driven key sentence extraction method to obtain the crucial content of the text and filter out noise. Subsequently, it integrates Bi-Encoder and Cross-Encoder to better capture semantic features and matching signals. Extensive experiments on several publicly available datasets demonstrate the effectiveness of our approach, compared with strong baselines. Furthermore, our model exhibits greater stability and accuracy in determining the matching relationship between documents describing the same event, which outperforms previously established approaches. The code is released at https://github.com/CSU-NLP-Group/FBC.

Distributed Machine Learning (DML) techniques are widely used to accelerate the training of large-scale machine learning models. However, during training iterations, gradients need to be frequently aggregated across multiple workers, resulting in communication bottleneck. To reduce the communication overhead of DML, several In-Network Aggregation (INA) protocols are proposed to reduce the volume of aggregation traffic by offloading aggregation functions into switches, thus alleviating network bottlenecks. Nevertheless, these protocols couple the congestion control of in-switch aggregator resources and link bandwidth resources, together with the straggler-oblivious manner in aggregator allocation, leading to low aggregation efficiency.

In this paper, we study the $k$-clustering problems with outliers in distributed setting. The current best results for the distributed $k$-center problem with outliers have quadratic local running time with communication cost dependent on the aspect ratio $\Delta$ of the given instance, which may constraint the scalability of the algorithms for handling large-scale datasets. To achieve better communication cost for the problem with faster local running time, we propose an inliers-recalling sampling method, which avoids guessing the optimal radius of the given instance, and can achieve a 4-round bi-criteria $(14(1+\epsilon),1+\epsilon)$-approximation with linear local running time in the data size and communication cost independent of the aspect ratio. To obtain a more practical algorithm for the problem, we propose another space-narrowing sampling method, which automatically adjusts the sample size to adapt to different outliers distributions on each machine, and can achieve a 2-round bi-criteria $(14(1+\epsilon),1+\epsilon)$-approximation with communication cost independent of the number of outliers. We show that, if the data points are randomly partitioned across machines, our proposed sampling-based methods can be extended to the $k$-median/means problems with outliers, and can achieve $(O(\frac{1}{\epsilon^2}),1+\epsilon)$-approximation with communication cost independent of the number of outliers. Empirical experiments suggest that the proposed 2-round distributed algorithms outperform other state-of-the-art algorithms.

Large-scale machine learning typically adopts distributed machine learning (DML) techniques to accelerate model training. Due to the large communication overhead, unfortu-nately, the phase of gradient aggregation has become the performance bottleneck for DML. To reduce traffic volume, several in-network aggregation (INA) transmission protocols are proposed to offload gradient aggregation function into the programmable switches. However, since existing INA transmission protocols use synchronous congestion control mechanism to drive each round of gradient aggregation, the straggling workers lead to long iteration time and significant performance degradation. To solve the above problem, we propose PA-ATP, a progress-aware INA transmission protocol, which adopts the progress-aware asynchronous congestion control. PA-ATP adjusts the sending rate in accordance with the transmission progress, allowing the straggling flow to grab more bandwidth than the leading flow and control the asynchronous degree of straggling job. We use a P4 programmable switch and a kernel-bypass protocol stack to implement PA-ATP. The results of testbed and large-scale NS3 simulations show that PA-ATP reduces training time by up to 62% compared to the state-of-the-art INA transmission protocols.

We present a $9^k\cdot n^{O(1)}$-time algorithm for the proper circular-arc vertex deletion problem, resolving an open problem of van 't Hof and Villanger [Algorithmica 2013] and Crespelle et al. [arXiv:2001.06867]. Our structural study also implies parameterized algorithms for modification problems toward proper Helly circular-arc graphs.

The local search methods have been widely used to solve the clustering problems. In practice, local search algorithms for clustering problems mainly adapt the single-swap strategy, which enables them to handle large-scale datasets and achieve linear running time in the data size. However, compared with multi-swap local search algorithms, there is a considerable gap on the approximation ratios of the single-swap local search algorithms. Although the current multi-swap local search algorithms provide small constant approximation, the proposed algorithms tend to have large polynomial running time, which cannot be used to handle large-scale datasets. In this paper, we propose a multi-swap local search algorithm for the $k$-means problem with linear running time in the data size. Given a swap size $t$, our proposed algorithm can achieve a $(50(1+\frac{1}{t})+\epsilon)$-approximation, which improves the current best result 509 (ICML 2019) with linear running time in the data size. Our proposed method, compared with previous multi-swap local search algorithms, is the first one to achieve linear running time in the data size. To obtain a more practical algorithm for the problem with better clustering quality and running time, we propose a sampling-based method which accelerates the process of clustering cost update during swaps. Besides, a recombination mechanism is proposed to find potentially better solutions. Empirical experiments show that our proposed algorithms achieve better performances compared with branch and bound solver (NeurIPS 2022) and other existing state-of-the-art local search algorithms on both small and large datasets.

GNN explanation method aims to identify an explanatory subgraph which contains the most informative components of the full graph. However, a major limitation of existing GNN explainers is that they are not robust to the structurally corrupted graphs, e.g., graphs with noisy or adversarial edges. On the one hand, existing GNN explainers mostly explore explanations based on either the raw graph features or the learned latent representations, both of which can be easily corrupted. On the other hand, the corruptions in graphs are irregular in terms of the structural properties, e.g., the size or connectivity of graphs, which makes the rigorous constraints used by previous GNN explainers unfeasible. To address these issues, we propose a robust GNN explainer called V-InfoR. Specifically, a robust graph representation extractor, which takes insights of variational inference, is proposed to infer the latent distribution of graph representations. Instead of directly using the corrupted raw features or representations of each single graph, we sample the graph representations from the inferred distribution for the downstream explanation generator, which can effectively eliminate the minor corruption. We next formulate the explanation exploration as a graph information bottleneck (GIB) optimization problem. As a more general method that does not need any rigorous structural constraints, our GIB-based method can adaptively capture both the regularity and irregularity of the severely corrupted graphs for explanation. Extensive evaluations on both synthetic and real-world datasets indicate that V-InfoR significantly improves the GNN explanation performance for the structurally corrupted graphs. Code and dataset are available at https://anonymous.4open.science/r/V-InfoR-EF88

Graph contrastive learning (GCL) has attracted rising research attention recently due to its effectiveness in self- supervised graph learning. A key step of GCL is to conduct data augmentation, based on which self-supervised learning is performed through the contrast between two augmented data views. Existing approaches generally generate the two data views from the original graph, which has been revealed to be less effective due to the lack of data diversity. Meanwhile, although the data augmentation methods and the contrastive modes have been extensively studied, the effect of hard negative samples (i.e.samples that are difficult to distinguish from an anchor node) on GCL is not fully explored. In this paper, we propose a novel complementary graph contrastive learning method boosted by adversarial hard negative sample generation. Specifically, we first construct a κNN graph as the complementary counterpart of the original graph in the semantic space. Then graph augmentation is conducted in both the semantic and topology spaces for the two complementary graphs to obtain two contrastive views with a larger data diversity. To facilitate the contrastive learning, an adversarial network named ADNet is also proposed to generate hard negative samples. The generated samples are more informative and challenging, and thus can further boost the learning performance. Extensive evaluations over the node classification task demonstrate that our proposal outperforms existing state-of-the-art GCL methods, and even exceeds supervised approaches. The code of this work is publicly available at https://github.com/sktsherlock/HNGCL-V1.

Abstract We study the parameterized complexity of winner determination problems for three prevalent k-committee selection rules, namely the minimax approval voting (MAV), the proportional approval voting (PAV), and the Chamberlin–Courant’s approval voting (CCAV). It is known that these problems are computationally hard. Although they have been studied from the parameterized complexity point of view with respect to several natural parameters, many of them turned out to be -hard or -hard. Aiming at obtaining plentiful fixed-parameter algorithms, we revisit these problems by considering more natural single parameters, combined parameters, and structural parameters.

Accumulating evidences demonstrate that circular RNA (circRNA) plays an important role in human diseases. Identification of circRNA-disease associations can help for the diagnosis of human diseases, while the traditional method based on biological experiments is time-consuming. In order to address the limitation, a series of computational methods have been proposed in recent years. However, few works have summarized these methods or compared the performance of them. In this paper, we divided the existing methods into three categories: information propagation, traditional machine learning and deep learning. Then, the baseline methods in each category are introduced in detail. Further, 5 different datasets are collected, and 14 representative methods of each category are selected and compared in the 5-fold, 10-fold cross-validation and the de novo experiment. In order to further evaluate the effectiveness of these methods, six common cancers are selected to compare the number of correctly identified circRNA-disease associations in the top-10, top-20, top-50, top-100 and top-200. In addition, according to the results, the observation about the robustness and the character of these methods are concluded. Finally, the future directions and challenges are discussed.

Emerging evidence has proved that circular RNAs (circRNAs) are implicated in pathogenic processes. They are regarded as promising biomarkers for diagnosis due to covalently closed loop structures. As opposed to traditional experiments, computational approaches can identify circRNA-disease associations at a lower cost. Aggregating multi-source pathogenesis data helps to alleviate data sparsity and infer potential associations at the system level. The majority of computational approaches construct a homologous network using multi-source data, but they lose the heterogeneity of the data. Effective methods that use the features of multi-source data are considered as a matter of urgency. In this paper, we propose a model (CDHGNN) based on edge-weighted graph attention and heterogeneous graph neural networks for potential circRNA-disease association prediction. The circRNA network, micro RNA network, disease network and heterogeneous network are constructed based on multi-source data. To reflect association probabilities between nodes, an edge-weighted graph attention network model is designed for node features. To assign attention weights to different types of edges and learn contextual meta-path, CDHGNN infers potential circRNA-disease association based on heterogeneous neural networks. CDHGNN outperforms state-of-the-art algorithms in terms of accuracy. Edge-weighted graph attention networks and heterogeneous graph networks have both improved performance significantly. Furthermore, case studies suggest that CDHGNN is capable of identifying specific molecular associations and investigating biomolecular regulatory relationships in pathogenesis. The code of CDHGNN is freely available at https://github.com/BioinformaticsCSU/CDHGNN.

Circular RNAs (circRNAs) are reverse-spliced and covalently closed RNAs. Their interactions with RNA-binding proteins (RBPs) have multiple effects on the progress of many diseases. Some computational methods are proposed to identify RBP binding sites on circRNAs but suffer from insufficient accuracy, robustness and explanation. In this study, we first take the characteristics of both RNA and RBP into consideration. We propose a method for discriminating circRNA-RBP binding sites based on multi-scale characterizing sequence and structure features, called CRMSS. For circRNAs, we use sequence ${k}\hbox{-}{mer}$ embedding and the forming probabilities of local secondary structures as features. For RBPs, we combine sequence and structure frequencies of RNA-binding domain regions to generate features. We capture binding patterns with multi-scale residual blocks. With BiLSTM and attention mechanism, we obtain the contextual information of high-level representation for circRNA-RBP binding. To validate the effectiveness of CRMSS, we compare its predictive performance with other methods on 37 RBPs. Taking the properties of both circRNAs and RBPs into account, CRMSS achieves superior performance over state-of-the-art methods. In the case study, our model provides reliable predictions and correctly identifies experimentally verified circRNA-RBP pairs. The code of CRMSS is freely available at https://github.com/BioinformaticsCSU/CRMSS.

MOTIVATION: Medical genomics faces significant challenges in interpreting disease phenotype and genetic heterogeneity. Despite the establishment of standardized disease phenotype databases, computational methods for predicting gene-phenotype associations still suffer from imbalanced category distribution and a lack of labeled data in small categories. RESULTS: To address the problem of labeled-data scarcity, we propose a self-supervised learning strategy for gene-phenotype association prediction, called SSLpheno. Our approach utilizes an attributed network that integrates protein-protein interactions and gene ontology data. We apply a Laplacian-based filter to ensure feature smoothness and use self-supervised training to optimize node feature representation. Specifically, we calculate the cosine similarity of feature vectors and select positive and negative sample nodes for reconstruction training labels. We employ a deep neural network for multi-label classification of phenotypes in the downstream task. Our experimental results demonstrate that SSLpheno outperforms state-of-the-art methods, especially in categories with fewer annotations. Moreover, our case studies illustrate the potential of SSLpheno as an effective prescreening tool for gene-phenotype association identification. AVAILABILITY AND IMPLEMENTATION: https://github.com/bixuehua/SSLpheno.

MOTIVATION: Oxford Nanopore sequencing has great potential and advantages in population-scale studies. Due to the cost of sequencing, the depth of whole-genome sequencing for per individual sample must be small. However, the existing single nucleotide polymorphism (SNP) callers are aimed at high-coverage Nanopore sequencing reads. Detecting the SNP variants on low-coverage Nanopore sequencing data is still a challenging problem. RESULTS: We developed a novel deep learning-based SNP calling method, NanoSNP, to identify the SNP sites (excluding short indels) based on low-coverage Nanopore sequencing reads. In this method, we design a multi-step, multi-scale and haplotype-aware SNP detection pipeline. First, the pileup model in NanoSNP utilizes the naive pileup feature to predict a subset of SNP sites with a Bi-long short-term memory (LSTM) network. These SNP sites are phased and used to divide the low-coverage Nanopore reads into different haplotypes. Finally, the long-range haplotype feature and short-range pileup feature are extracted from each haplotype. The haplotype model combines two features and predicts the genotype for the candidate site using a Bi-LSTM network. To evaluate the performance of NanoSNP, we compared NanoSNP with Clair, Clair3, Pepper-DeepVariant and NanoCaller on the low-coverage (∼16×) Nanopore sequencing reads. We also performed cross-genome testing on six human genomes HG002-HG007, respectively. Comprehensive experiments demonstrate that NanoSNP outperforms Clair, Pepper-DeepVariant and NanoCaller in identifying SNPs on low-coverage Nanopore sequencing data, including the difficult-to-map regions and major histocompatibility complex regions in the human genome. NanoSNP is comparable to Clair3 when the coverage exceeds 16×. AVAILABILITY AND IMPLEMENTATION: https://github.com/huangnengCSU/NanoSNP.git. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: A single gene may yield several isoforms with different functions through alternative splicing. Continuous efforts are devoted to developing machine-learning methods to predict isoform functions. However, existing methods do not consider the relevance of each feature to specific functions and ignore the noise caused by the irrelevant features. In this case, we hypothesize that constructing a feature selection framework to extract the function-relevant features might help improve the model accuracy in isoform function prediction. RESULTS: In this article, we present a feature selection-based approach named IsoFrog to predict isoform functions. First, IsoFrog adopts a reversible jump Markov Chain Monte Carlo (RJMCMC)-based feature selection framework to assess the feature importance to gene functions. Second, a sequential feature selection procedure is applied to select a subset of function-relevant features. This strategy screens the relevant features for the specific function while eliminating irrelevant ones, improving the effectiveness of the input features. Then, the selected features are input into our proposed method modified domain-invariant partial least squares, which prioritizes the most likely positive isoform for each positive MIG and utilizes diPLS for isoform function prediction. Tested on three datasets, our method achieves superior performance over six state-of-the-art methods, and the RJMCMC-based feature selection framework outperforms three classic feature selection methods. We expect this proposed methodology will promote the identification of isoform functions and further inspire the development of new methods. AVAILABILITY AND IMPLEMENTATION: IsoFrog is freely available at https://github.com/genemine/IsoFrog.

MOTIVATION: Hi-C technology has been the most widely used chromosome conformation capture (3C) experiment that measures the frequency of all paired interactions in the entire genome, which is a powerful tool for studying the 3D structure of the genome. The fineness of the constructed genome structure depends on the resolution of Hi-C data. However, due to the fact that high-resolution Hi-C data require deep sequencing and thus high experimental cost, most available Hi-C data are in low-resolution. Hence, it is essential to enhance the quality of Hi-C data by developing the effective computational methods. RESULTS: In this work, we propose a novel method, so-called DFHiC, which generates the high-resolution Hi-C matrix from the low-resolution Hi-C matrix in the framework of the dilated convolutional neural network. The dilated convolution is able to effectively explore the global patterns in the overall Hi-C matrix by taking advantage of the information of the Hi-C matrix in a way of the longer genomic distance. Consequently, DFHiC can improve the resolution of the Hi-C matrix reliably and accurately. More importantly, the super-resolution Hi-C data enhanced by DFHiC is more in line with the real high-resolution Hi-C data than those done by the other existing methods, in terms of both chromatin significant interactions and identifying topologically associating domains. AVAILABILITY AND IMPLEMENTATION: https://github.com/BinWangCSU/DFHiC.

MOTIVATION: Single-cell RNA sequencing (scRNA-seq) offers a powerful tool to dissect the complexity of biological tissues through cell sub-population identification in combination with clustering approaches. Feature selection is a critical step for improving the accuracy and interpretability of single-cell clustering. Existing feature selection methods underutilize the discriminatory potential of genes across distinct cell types. We hypothesize that incorporating such information could further boost the performance of single cell clustering. RESULTS: We develop CellBRF, a feature selection method that considers genes' relevance to cell types for single-cell clustering. The key idea is to identify genes that are most important for discriminating cell types through random forests guided by predicted cell labels. Moreover, it proposes a class balancing strategy to mitigate the impact of unbalanced cell type distributions on feature importance evaluation. We benchmark CellBRF on 33 scRNA-seq datasets representing diverse biological scenarios and demonstrate that it substantially outperforms state-of-the-art feature selection methods in terms of clustering accuracy and cell neighborhood consistency. Furthermore, we demonstrate the outstanding performance of our selected features through three case studies on cell differentiation stage identification, non-malignant cell subtype identification, and rare cell identification. CellBRF provides a new and effective tool to boost single-cell clustering accuracy. AVAILABILITY AND IMPLEMENTATION: All source codes of CellBRF are freely available at https://github.com/xuyp-csu/CellBRF.

MOTIVATION: Cancer heterogeneity drastically affects cancer therapeutic outcomes. Predicting drug response in vitro is expected to help formulate personalized therapy regimens. In recent years, several computational models based on machine learning and deep learning have been proposed to predict drug response in vitro. However, most of these methods capture drug features based on a single drug description (e.g. drug structure), without considering the relationships between drugs and biological entities (e.g. target, diseases, and side effects). Moreover, most of these methods collect features separately for drugs and cell lines but fail to consider the pairwise interactions between drugs and cell lines. RESULTS: In this paper, we propose a deep learning framework, named MSDRP for drug response prediction. MSDRP uses an interaction module to capture interactions between drugs and cell lines, and integrates multiple associations/interactions between drugs and biological entities through similarity network fusion algorithms, outperforming some state-of-the-art models in all performance measures for all experiments. The experimental results of de novo test and independent test demonstrate the excellent performance of our model for new drugs. Furthermore, several case studies illustrate the rationality for using feature vectors derived from drug similarity matrices from multisource data to represent drugs and the interpretability of our model. AVAILABILITY AND IMPLEMENTATION: The codes of MSDRP are available at https://github.com/xyzhang-10/MSDRP.

This Supplement issue, presents five research articles which are distributed, mainly due to the subject they address, from the 8th International Work-Conference on Bioinformatics and Biomedical Engineering (IWBBIO 2020), which was held on line, during September, 30th-2nd October, 2020. These contributions have been chosen because of their quality and the importance of their findings. Those contributions were then invited to participate in this supplement for the following journals of BMC: BMC Bioinformatics and BMC Genomics. In the present Editorial in BMC journal, we summarize the contributions that provide a clear overview of the thematic areas covered by the IWBBIO conference, ranging from theoretical/review aspects to real-world applications of bioinformatic and biomedical engineering.

Unattended wireless sensor networks (UWSNs), in which mobile sinks (MSs) are responsible for the data collection, and which are vulnerable to multiple attacks. For example, the adversaries can initiate MS replication attack after compromising a large number of sensor nodes (SNs). To resist such attacks, some scholars have proposed some schemes. In these schemes, it is assumed that MSs are equipped with tamper resistance hardware, so they are pre-distributed most of keys of a key pool for achieving authentication with SNs. However, in outdoor and even sensitive areas, tamper-resistant hardware is not always absolutely safe. Once MSs are compromised, networks become insecure. In this paper, we propose a secure model. The model contains three types of nodes, namely SNs, MSs, and a base station (BS). MSs are responsible for collecting the data encrypted by SNs and forwarding it to BS; BS is responsible for decrypting and analyzing the collected data. During the data collection process, an MS can collect an SN's data only after being authenticated by the SN using the pre-distribution key information. But it cannot decrypt the collected data. In this model, the adversaries can induce a new type of false data injection attack. In the attack, the replicated MSs impersonate uncompromised SNs to send false data to BS by using the compromised key information. If BS accepts a large amount of false data, it will make a wrong judgment. Analysis and simulation show that the proposed model has excellent resistance against MS replication attacks and false data injection attacks by setting appropriate parameter values.

Abstract Datacenter applications generate diverse flows, including deadline flows and non‐deadline flows. The deadline flows require to complete within strict deadline, while non‐deadline flows seek a shorter flow completion time. The state‐of‐the‐art deadline‐aware methods either transmit deadline flows with best‐effort at high priority, resulting in the starvation of non‐deadline flows, or blindly restrict the sending rates of deadline flows, leading to a high deadline missing ratio. To meet the different requirements of mix‐flows, a novel traffic‐aware rate control (TRC) method is proposed. TRC dynamically adjusts the sending rates of deadline flows according to their deadlines and the predicted future traffic patterns. If the intense competition is predicted among deadline flows, TRC will adopt a more aggressive manner to transmit the current deadline flows to avoid bandwidth contention in the future, reducing the deadline missing ratio. Otherwise, TRC will conservatively transmit deadline flows and complete these flows near their respective deadlines, relinquishing the excess bandwidth to non‐deadline flows. Meanwhile, TRC schedules non‐deadline flows in accordance with their sizes, minimizing the average FCT. The performance of TRC in large‐scale scenarios is evaluated through NS2 simulations. The test results show that TRC reduces the deadline missing ratio of deadline flows and the FCT of non‐deadline flows by up to 69.5% and 78.7% compared to the state‐of‐the‐art deadline‐aware schemes, respectively.

Arrhythmia is an abnormal heart rhythm, a common clinical problem in cardiology. Long-term or severe arrhythmia may lead to stroke and sudden cardiac death. The electrocardiogram (ECG) is the most commonly used tool to diagnose arrhythmia. However, the traditional diagnosis relies on experts for manual interpretation, which is time-consuming and laborious. In recent years, many automatic arrhythmia detection methods have emerged due to advancements in deep learning. These methods can reduce manual intervention and improve diagnostic efficiency. However, extracting useful features from raw ECG signals for arrhythmia detection is still challenging due to the low frequency of ECG signals and noise distribution. In this paper, we propose a novel hidden attention residual network (HA-ResNet) for automated arrhythmia classification. In this model, the one-dimensional ECG signals are first converted into two-dimensional images and fed into an embedding layer to obtain the relevant shallow features in ECG. Then, a hidden attention layer combining Squeeze-and-Excitation (SE) block and Bidirectional Convolutional LSTM (BConvLSTM) is used to further capture the deep Spatio-temporal features. We evaluate our HA-ResNet on two public datasets and achieve F1 scores of 96.0%, 96.7%, and 87.6% on 2s segments, 5s segments, and 10s segments, respectively, which significantly outperform the existing state-of-the-art approaches. The experimental results demonstrate the effectiveness and generalization of our method.

Nuclei segmentation is an essential step in DNA ploidy analysis by image-based cytometry (DNA-ICM) which is widely used in cytopathology and allows an objective measurement of DNA content (ploidy). The routine fully supervised learning-based method requires often tedious and expensive pixel-wise labels. In this paper, we propose a novel weakly supervised nuclei segmentation framework which exploits only sparsely annotated bounding boxes, without any segmentation labels. The key is to integrate the traditional image segmentation and self-training into fully supervised instance segmentation. We first leverage the traditional segmentation to generate coarse masks for each box-annotated nucleus to supervise the training of a teacher model, which is then responsible for both the refinement of these coarse masks and pseudo labels generation of unlabeled nuclei. These pseudo labels and refined masks along with the original manually annotated bounding boxes jointly supervise the training of student model. Both teacher and student share the same architecture and especially the student is initialized by the teacher. We have extensively evaluated our method with both our DNA-ICM dataset and public cytopathological dataset. Without bells and whistles, our method outperforms all existing weakly supervised entries on both datasets. Code and our DNA-ICM dataset are publicly available at https://github.com/CVIU-CSU/Weakly-Supervised-Nuclei-Segmentation.

Topologically associating domains (TADs) are local chromatin interaction domains, which have been shown to play an important role in gene expression regulation. TADs were originally discovered in the investigation of 3D genome organization based on High-throughput Chromosome Conformation Capture (Hi-C) data. Continuous considerable efforts have been dedicated to developing methods for detecting TADs from Hi-C data. Different computational methods for TADs identification vary in their assumptions and criteria in calling TADs. As a consequence, the TADs called by these methods differ in their similarities and biological features they are enriched in. In this work, we performed a systematic comparison of twenty-six TAD callers. We first compared the TADs and gaps between adjacent TADs across different methods, resolutions, and sequencing depths. We then assessed the quality of TADs and TAD boundaries according to three criteria: the decay of contact frequencies over the genomic distance, enrichment and depletion of regulatory elements around TAD boundaries, and reproducibility of TADs and TAD boundaries in replicate samples. Last, due to the lack of a gold standard of TADs, we also evaluated the performance of the methods on synthetic datasets. We discussed the key principles of TAD callers, and pinpointed current situation in the detection of TADs. We provide a concise, comprehensive, and systematic framework for evaluating the performance of TAD callers, and expect our work will provide useful guidance in choosing suitable approaches for the detection and evaluation of TADs.

Single cell RNA sequencing (scRNA-seq) provides a powerful approach for profiling transcriptomes at single cell resolution. An essential application of scRNA-seq is the discovery of cell types with the aid of clustering analysis. Currently, existing single cell clustering methods are exclusively based on gene-level expression data, without considering alternative splicing information. It has been shown that alternative splicing has an important influence on biological processes such as cell differentiation and cell cycle. We therefore hypothesize that adding information about alternative splicing may help enhance single cell clustering. This motivates us to develop a way to integrate isoform-level expression and gene-level expression. We report an approach to enhance single cell clustering by integrating isoform-level expression through orthogonal projection. First, we construct an orthogonal projection matrix based on gene expression data. Second, isoforms are projected to the gene space to remove the redundant information between them. Third, isoform selection is performed based on the residual of the projected expression and the selected isoforms are combined with gene expression data for subsequent clustering. We applied our method to sixteen scRNA-seq datasets. We find that alternative splicing contains differential information among cell types and can be integrated to enhance single cell clustering. Compared with using only gene-level expression data, the integration of isoform-level expression leads to better clustering performances for most of the datasets. The integration of isoform-level expression also has potential in the detection of novel cell subgroups. Our study shows that integrating isoform and gene-level expression is a promising way to improve single cell clustering. The IsoCell R package is freely available at both Github (https://github.com/genemine/IsoCell) and Zenodo (https://zenodo.org/record/4395707).

Alzheimer's disease (AD) is the most common neurodegenerative disease. More and more evidence show that DNA methylation is closely related to the pathological mechanism of AD. Many AD-associated differentially methylated genes, regions and CpG sites have been identified in recent researches, which may have great potential in clinical research. However, there is no dedicated database to collect AD-related differential methylation up to now. To provide a reference to researchers, we design a database named ADmeth by manually curating relevant articles, which contains a total of 16,709 AD-related differentially methylated items identified from different brain regions and different cell types in the blood, involving 209 genes, 2,229 regions and 14,271 CpG sites. The ADmeth database provides user-friendly pages to search, submit and download data. We hope that the ADmeth database can facilitate researchers to select candidate AD-associated methylation markers in revealing the pathological mechanism of AD and promote the cell-free DNA based non-invasive diagnosis of AD. The ADmeth database is available at http://www.biobdlab.cn/ADmeth.

The Anatomical Therapeutic Chemical (ATC) classification system, designated by the World Health Organization Collaborating Center (WHOCC), has been widely used in drug screening, repositioning, and similarity research. The ATC classification system assigns different codes to drugs according to the organ or system on which they act and/or their therapeutic and chemical characteristics. Correctly identifying the potential ATC codes for drugs can accelerate drug development and reduce the cost of experiments. Several classifiers have been proposed in this regard. However, they lack of ability to learn basic features from sparsely known drug-ATC code associations. Therefore, there is an urgent need for novel computational methods to precisely predict potential drug-ATC code associations in multiple levels of the ATC classification system based on known associations between drugs and ATC codes. In this paper, we provide a novel end-to-end model, so-called RNPredATC, to predict potential drug-ATC code associations in five ATC classification levels. RNPredATC can extract dense feature vectors from sparsely known drug-ATC code associations and reduce the impact from the degradation problem by a novel deep residual learning. We extensively compare our method with some state-of-the-art methods, including NetPredATC, SPACE, and some multi-label-based methods. Our experimental results show that RNPredATC achieves better performances in five-fold and ten-fold cross validations. Furthermore, the visualization analysis of hidden layers and case studies of predicted associations at the fifth ATC classification level confirm that RNPredATC can effectively identify the potential ATC codes of drugs.

The identification of drug-target relations (DTRs) is substantial in drug development. A large number of methods treat DTRs as drug-target interactions (DTIs), a binary classification problem. The main drawback of these methods are the lack of reliable negative samples and the absence of many important aspects of DTR, including their dose dependence and quantitative affinities. With increasing number of publications of drug-protein binding affinity data recently, DTRs prediction can be viewed as a regression problem of drug-target affinities (DTAs) which reflects how tightly the drug binds to the target and can present more detailed and specific information than DTIs. The growth of affinity data enables the use of deep learning architectures, which have been shown to be among the state-of-the-art methods in binding affinity prediction. Although relatively effective, due to the black-box nature of deep learning, these models are less biologically interpretable. In this study, we proposed a deep learning-based model, named AttentionDTA, which uses attention mechanism to predict DTAs. Different from the models using 3D structures of drug-target complexes or graph representation of drugs and proteins, the novelty of our work is to use attention mechanism to focus on key subsequences which are important in drug and protein sequences when predicting its affinity. We use two separate one-dimensional Convolution Neural Networks (1D-CNNs) to extract the semantic information of drug's SMILES string and protein's amino acid sequence. Furthermore, a two-side multi-head attention mechanism is developed and embedded to our model to explore the relationship between drug features and protein features. We evaluate our model on three established DTA benchmark datasets, Davis, Metz, and KIBA. AttentionDTA outperforms the state-of-the-art deep learning methods under different evaluation metrics. The results show that the attention-based model can effectively extract protein features related to drug information and drug features related to protein information to better predict drug target affinities. It is worth mentioning that we test our model on IC50 dataset, which provides the binding sites between drugs and proteins, to evaluate the ability of our model to locate binding sites. Finally, we visualize the attention weight to demonstrate the biological significance of the model. The source code of AttentionDTA can be downloaded from https://github.com/zhaoqichang/AttentionDTA_TCBB.

Identifying the function of therapeutic peptides is an important issue in the development of novel drugs. To reduce the time and labor costs required to identify therapeutic peptides, computational methods are increasingly required. However, most of the existing peptide therapeutic function prediction models are used for predicting a single therapeutic function, ignoring the fact that a bioactive peptide might simultaneously consist of multi-activities. Furthermore, in the few existing multi-label classification models, the feature extraction procedures are still rough. We propose a multi-label framework, called SCN-MLTPP, with a stacked capsule network for predicting the therapeutic properties of peptides. Instead of using peptide sequence vectors alone, SCN-MLTPP extracts different view representation vectors from the therapeutic peptides and learns the contributions of different views to the properties of therapeutic peptides based on the dynamic routing mechanism. Benchmarking results show that as compared with existing multi-label predictors, SCN-MLTPP achieves better and more robust performance for different peptides. In addition, some visual analyses and case studies also demonstrate the model can reliably capture features from multi-view data and predict different peptides.

Drug discovery and drug repurposing often rely on the successful prediction of drug-target interactions (DTIs). Recent advances have shown great promise in applying deep learning to drug-target interaction prediction. One challenge in building deep learning-based models is to adequately represent drugs and proteins that encompass the fundamental local chemical environments and long-distance information among amino acids of proteins (or atoms of drugs). Another challenge is to efficiently model the intermolecular interactions between drugs and proteins, which plays vital roles in the DTIs. To this end, we propose a novel model, GIFDTI, which consists of three key components: the sequence feature extractor (CNNFormer), the global molecular feature extractor (GF), and the intermolecular interaction modeling module (IIF). Specifically, CNNFormer incorporates CNN and Transformer to capture the local patterns and encode the long-distance relationship among tokens (atoms or amino acids) in a sequence. Then, GF and IIF extract the global molecular features and the intermolecular interaction features, respectively. We evaluate GIFDTI on six realistic evaluation strategies and the results show it improves DTI prediction performance compared to state-of-the-art methods. Moreover, case studies confirm that our model can be a useful tool to accurately yield low-cost DTIs. The codes of GIFDTI are available at https://github.com/zhaoqichang/GIFDTI.

Since most flows are short-lived in data center networks, fast convergence becomes very important to help the short flows effectively utilize high bandwidth. Though current explicit feedback-based transport control protocols (TCPs) provide fast convergence via fine-grained congestion information from customized switches, they unavoidably incur large traffic overhead for widely existing small packets in data center applications, resulting in suboptimal network efficiency. To solve this issue, we propose a datacenter TCP based onDifferentialExplicitCongestionNotification, called DECN, to achieve fast convergence without any traffic overhead. Specifically, DECN feeds rate difference between the target and current rate back to the source by using multiple consecutive packets. Besides, we propose an enhanced version DECN* which obtains the optimal number of consecutive packets according to the packet loss rate. The experimental results of NS2 simulation and testbed implementation show that DECN and its enhanced version DECN* achieve comparable fast convergence as XCP without incurring any extra feedback overhead. Compared with the state-of-the-art explicit feedback-based TCPs, they reduce the flow completion time by up to 34% in typical data center applications.

Distributed key-value stores provide the Multiget API, where many key-value operations are batched together, to meet the parallel requirement of applications. Correspondingly, reducing the latency of Multigets is crucial for the responsiveness of the distributed key-value stores. The latency of a Multiget depends on both which replica server its key-value operations are scheduled to, i.e., the replica selection for each key-value operation, and when these key-value operations are served, i.e., the scheduling of the service sequence at different replica servers. Existing solutions solely focus on either one of them and accordingly lead to suboptimal latency for Multigets. To address these issues, this paper proposes an Adaptive Multiget Scheduling (AMS) algorithm in this paper, and specifically, our AMS re-architectures the framework to remove the conflict between replica selection and service sequence scheduling in the existing solution Rein. Based on the new framework, a sophisticated replica selection method is designed. Furthermore, AMS guides both replica selection and service sequence scheduling by the piggybacked information of replica servers, being adaptive to the heterogeneous time-varying server performance. Consequently, AMS can respectively reduce the median,$95^{th}$, and$99^{th}$percentile latencies of Multigets by a factor of 4, 3.1, and 1.86 compared to the default FIFO algorithm and significantly outperforms Rein.

In recent years, receiver-driven transport protocols have been proposed to use proactive congestion control to meet the stringent latency requirements of large-scale applications in data center. However, the receiver-driven proposals face the challenges brought by network dynamic. First, when the bursty flows start, the aggressive and blind line-rate transmission in the first RTT easily leads to persistent queue backlog. Second, when some flows finish transmissions, the remaining ones cannot increase their sending rates to seize the available bandwidth. To address these problems, this article presents a new receiver-driven congestion control design, called REN, which uses the under- and over-utilization notifications from switch to handle the dynamic traffic. With the aid of explicit feedback, REN alleviates the traffic burstiness due to aggressive start, mitigates the conservativeness in utilizing available bandwidth, and still retains the receiver-driven feature to achieve ultra-low latency. We implement the prototype of REN using DPDK. The experimental results of real testbed and large-scale NS2 simulation show that REN effectively reduces the average flow completion time (AFCT) by up to 68% over the state-of-the-art receiver-driven transmission schemes.

The three-dimensional deployment of Unmanned Aerial Vehicles (UAVs) has attracted extensive attention, especially for the Internet of Vehicles (IoV) in an emergency or to help the overloaded edge servers in traffic peaks. However, most existing works assume a two-dimensional road to simplify the design and modeling, while ignoring the interchange bridges scenario. In this scenario, UAVs deployment will face new challenges: the line-of-sight (LoS) transmission between the vehicles and UAVs is weakened due to the occlusion of the bridge body and vehicle movement. Meanwhile, energy consumption and the quantity of UAVs also need to be considered. In this paper, we propose an energy-aware 3D-deployment of UAVs, named 3D-UAV, to guarantee a high uplink rate with a minimized number of UAVs in IoV with Highway Interchange. First, considering the channel gain over bridges, 3D-UAV divides vehicles into several clusters. In each time slot, the number of clusters is iteratively optimized. Based on the clustering result, the flight altitude of the UAV is optimized in a stochastic gradient ascent (SGA) way aiming at maximizing the average uplink rate of transmission. Numerical results show that the proposed 3D-UAV can cover all vehicles on the highway interchange with the number of UAVs close to the theoretical lower bound. Meanwhile, it outperforms SOA, DRL, and HOLD methods in terms of the uplink rate and energy.

Greetings and welcome to the fifth issue of IEEE Transactions on Computational Social Systems (TCSS) for 2023. This edition presents a collection of 55 diverse regular articles that illuminate various facets of the interaction between computer technology and society.

The autonomous exploration task we consider requires Unmanned Aerial Vehicles (UAVs) to actively navigate through unknown environments with the goal of fully perceiving and mapping the environments. Some existing exploration strategies suffer from rough cost budgets, ambiguous Information Gain (IG), and unnecessary backtracking exploration caused by Fragmented Regions (FRs). In our work, a hierarchical spatio-temporal-aware exploration framework is proposed to alleviate these problems. At the local exploration level, the Asymmetrical Traveling Salesman Problem (ATSP) is solved by comprehensively considering exploration time, IG, and heading consistency to avoid blindly exploring. Specifically, the exploration time is reasonably budgeted by fast marching in an artificial potential field. Meanwhile, a transformer-based map occupancy predictor is designed to assist in IG calculation by imagining spatial clues out of the Field of View (FoV), facilitating the prescient exploration. We verify that our local exploration is effective in alleviating the unnecessary back-and-forth movements caused by FRs and the interference of potential obstacle occlusion on the IG calculation. At the global exploration level, the classical Next Best View Points (NBVP) are generalized to Next Best Sub-Regions (NBSR) to choose informative sub-regions for further forward-looking exploration based on a well-designed utility function. Safe flight paths and dynamically feasible trajectories are reasonably generated throughout the exploration process by fast marching and B-spline curve optimization. Comparative simulations and benchmark tests demonstrate that our proposed exploration strategy is quite competitive in terms of exploration path length, total exploration time, and exploration ratio.

Medical image segmentation is indispensable for diagnosis and prognosis of many diseases. To improve the segmentation performance, this study proposes a new 2D body and edge aware network with multi-scale short-term concatenation for medical image segmentation. Multi-scale short-term concatenation modules which concatenate successive convolution layers with different receptive fields, are proposed for capturing multi-scale representations with fewer parameters. Body generation modules with feature adjustment based on weight map computing via enlarging the receptive fields, and edge generation modules with multi-scale convolutions using Sobel kernels for edge detection, are proposed to separately learn body and edge features from convolutional features in decoders, making the proposed network be body and edge aware. Based on the body and edge modules, we design parallel body and edge decoders whose outputs are fused to achieve the final segmentation. Besides, deep supervision from the body and edge decoders is applied to ensure the effectiveness of the generated body and edge features and further improve the final segmentation. The proposed method is trained and evaluated on six public medical image segmentation datasets to show its effectiveness and generality. Experimental results show that the proposed method achieves better average Dice similarity coefficient and 95% Hausdorff distance than several benchmarks on all used datasets. Ablation studies validate the effectiveness of the proposed multi-scale representation learning modules, body and edge generation modules and deep supervision.

Cervical abnormal cell detection is a challenging task as the morphological discrepancies between abnormal and normal cells are usually subtle. To determine whether a cervical cell is normal or abnormal, cytopathologists always take surrounding cells as references to identify its abnormality. To mimic these behaviors, we propose to explore contextual relationships to boost the performance of cervical abnormal cell detection. Specifically, both contextual relationships between cells and cell-to-global images are exploited to enhance features of each region of interest (RoI) proposal. Accordingly, two modules, dubbed as RoI-relationship attention module (RRAM) and global RoI attention module (GRAM), are developed and their combination strategies are also investigated. We establish a strong baseline by using Double-Head Faster R-CNN with a feature pyramid network (FPN) and integrate our RRAM and GRAM into it to validate the effectiveness of the proposed modules. Experiments conducted on a large cervical cell detection dataset reveal that the introduction of RRAM and GRAM both achieves better average precision (AP) than the baseline methods. Moreover, when cascading RRAM and GRAM, our method outperforms the state-of-the-art (SOTA) methods. Furthermore, we show that the proposed feature-enhancing scheme can facilitate image- and smear-level classification.

Graph classification has been widely used for knowledge discovery in numerous practical application scenarios, such as social networks and protein-protein interaction networks. Recently, Graph Neural Networks (GNNs), which generalize deep neural networks to graph-structured data, have drawn considerable attention and achieved state-of-the-art performance in graph classification. However, existing GNN models mainly focus on capturing the information of immediate or first-order neighboring nodes within a single layer. The graph substructure and substructure interaction, that play an important role in learning graph representations, are usually overlooked. In this paper, we propose a Substructure Assembling Graph Attention Network (SA-GAT) to extract graph features and improve the performance of graph classification. SA-GAT is able to fully explore higher-order substructure information hidden in graphs by a core module called Substructure Interaction Attention (SIA). Theoretically, we have also proved that SA-GAT satisfies the graph isomorphism theory of graph neural network design, which is that the network should map isomorphic graphs to the same representation and output the same prediction. Extensive experimental results on multiple real-world graph classification datasets demonstrate that the proposed SA-GAT outperforms the state-of-the-art methods including graph kernels and graph neural networks.

Lip reading can help people with speech disorders to communicate with others and provide them with a new channel to interact with the world. In this paper, we design and implementHearMe, an accurate and real-time lip-reading system built on commercial RFID devices. HearMe can be used to accurately recognize different words in a pre-defined vocabulary without limitations in light conditions and can be used in multiple user scenarios by leveraging RFID's ability in identifying different users. We design an effective data collection strategy to well capture the tiny and complex signal patterns caused by mouth motion and propose a set of algorithms to extract signal profiles related to mouth motions and mitigate interference factors like multi-path. A carefully designed set of features, including time-domain statistical features and frequency-domain features, are then extracted from the signal to lift the recognition accuracy at the word level. To reduce training costs when the model is used in a new environment, a transfer-learning-based approach is adopted to enhance the robustness of the model in cross-environment scenarios. Experimental results show that HearMe detects speaking actions of the user with an accuracy higher than 0.95 and recognizes different words in a 20-words vocabulary with an average accuracy higher than 0.88. Moreover, the latency of HearMe ($\sim$150ms) is nearly two orders of magnitude less than traditional approaches, making it applicable to practical scenarios that require real-time lip reading.

Gesture recognition based on radio frequency identification (RFID) has attracted much research attention in recent years. Most existing RFID-based gesture recognition approaches use signal profile matching to distinguish different gestures, which incur large recognition latency and fail to support real-time applications. In this paper, we design and implement ReActor, a real-time and accurate gesture recognition system that recognizes a user's gestures with low latency and high accuracy even when the gestures'speed varies. ReActor combines the time-domain statistical features and the frequency-domain features to precisely represent the signal profile corresponding to different gestures. To maintain high accuracy across different environments, we preprocess the signals to remove reflection signals from surrounding objects and use only the signals related to gestures to train the classifier. Moreover, we train a classifier to predict the speed of the gesture and feed the extracted features to different classifiers according to the speed. We implement ReActor and evaluate its performance in different scenarios. Experimental results show that ReActor achieves an average accuracy of 97.2% in recognizing 18 different gestures with an average latency of 72 ms, more than two orders of magnitude faster than approaches based on profile template matching.

Most commercial players adopt adaptive bitrate (ABR) algorithms to dynamically decide each chunk's bitrate based on the perceived network bandwidth and buffer occupancy. However, current ABR algorithms are agnostic of audio bitrate selection since they deem it has negligible influence on video bitrate selection due to small size of audio chunks. Nevertheless, with the development of audio technologies, the bitrate of audio content increases dramatically in recent years. Thus, inappropriate audio selection can significantly affect video selection and deteriorate the viewing experience. To tackle these inefficiencies, we propose a deepReinforcement learning-based ABR algorithm that takesAudio andVideo quality into account (RAV) to circumvent a series of suboptimal performances, like low playback quality, frequent playback interruptions, poor playback smoothness, and undesirable combinations of video and audio chunks. Furthermore, RAV trains a neural network model that automatically outputs the bitrates for future audio and video chunks without relying on any presumptions about the environment, achieving good robustness to a broad spectrum of conditions. By conducting trace-driven and real-world experiments, we demonstrate that RAV significantly ameliorates the average overall viewing quality by 37.96%-118.20% over the state-of-the-art ABR algorithms. In addition, we also conduct subjective experiments by inviting 32 volunteers, and 27/32 users strongly agree that RAV provides them a better viewing experience than existing ABR solutions.

Video traffic has experienced an exponential increase in current years due to the growing ubiquity of mobile equipment and the constant network improvement. Most commercial players employ adaptive bitrate (ABR) algorithms to dynamically choose bitrate for each chunk based on perceived network capacity and buffer occupancy. Unluckily, even though improving the quality of chunks with dynamic scenes can achieve more QoE gain than static scenes, current ABR algorithms usually strive to maximize the average bitrate instead of perceptual quality, leading to the QoE degradation. To overcome this obstacle, we introduce a dynamic-chunk quality-aware adaptive bitrate algorithm through apprenticeship learning called DAVS (Dynamic-chunk qualityAwareVideoStreaming), where higher quality is selected for the dynamic chunks without reducing the quality of static chunks extravagantly. Furthermore, we take the user’s viewing preference into account to make DAVS adapt to the QoE diversity. The experimental results demonstrate that DAVS ameliorates the quality of dynamic chunks and significantly enhances the QoE compared with several representative ABR algorithms.

Existing reactive or proactive congestion control protocols are hard to simultaneously achieve ultra-low latency and high link utilization across all workloads ranging from delay-sensitive flows to bandwidth-hungry ones in datacenter networks. We present an Anti-ECN (Explicit Congestion Notification) Marking Receiver-driven Transport protocol called AMRT, which achieves both near-zero queueing delay and full link utilization by reasonably increasing sending rate in the case of under-utilization. Specifically, switches mark the ECN bit of data packets once detecting spare bandwidth. When receiving the anti-ECN marked packet, the receiver generates the corresponding marked grant to trigger more data packets. The testbed and simulation experiments show that AMRT effectively reduces the average flow completion time (AFCT) by up to 42% and improves the link utilization by up to 38% over the state-of-the-art receiver-driven transmission schemes.

Mobile Edge Computing (MEC) has attracted attention for its short-range and low-latency computing services for the Internet of Vehicles (IoV). However, in the Vehicle-RSU-MEC environment, incomplete task migration is a problem when Vehicle Users(VUs) are moving at high speeds around the Road Side Units (RSUs). Additionally, during peak traffic periods or road congestion, completed tasks compete for MEC server computing resources, leading to load imbalance. Therefore, making reasonable migration and offloading decisions is an important challenge. To address this challenge, the paper proposes a Cooperative Offloading strategy to jointly optimize offloading Decisions and Allocation of computing resources (CODA) step by step. First, to solve the problem of incomplete task migration, CODA proposed a Greedy-Based Task Completion Migration (GBTCM) algorithm. The algorithm calculates the required RSU set for each task to achieve complete migration, and greedily searches for the optimal migration target in the corresponding set to reduce task transmission latency. Second, after completing the task migration, CODA proposed a Distance-Based Computing Resource Allocation (DBCRA) algorithm to achieve load balancing for MEC servers. The algorithm prioritizes distance and finds MEC servers with sufficient computing resources to achieve better load balancing performance. Experimental results have shown that CODA is a low-complexity algorithm applicable to IoV, which can make reasonable and rapid decisions for task migration and offloading. Compared to three other benchmarks, CODA exhibits higher effectiveness and superiority.

Identifying heavy flows is essential for network management. However, it is challenging to detect heavy flow quickly and accurately under the highly dynamic traffic and rapid growth of network capacity. Existing heavy flow detection schemes can make a trade-off in efficiency, accuracy and speed. However, these schemes still require memory large enough to obtain acceptable performance. To address this issue, we propose ChainSketch, which has the advantages of good memory efficiency, high accuracy and fast detection. Specifically, ChainSketch uses the selective replacement strategy to mitigate the over-estimation issue. Meanwhile, ChainSketch utilizes the hash chain and compact structure to improve memory efficiency. We implement the ChainSketch on OVS platform, P4-based testbed and large-scale simulations to process heavy hitter and heavy changer detection. The results of trace-driven tests show that, ChainSketch greatly improves the F1-score by up to$3.43\times $compared with the state-of-the-art solutions especially for small memory.

Datacenter networks provide large bisection bandwidth by load balancing traffic over rich parallel paths in multi-rooted tree topologies. Nevertheless, production datacenters operate under various path diversities caused by traffic dynamics, hardware failures and heterogeneous switching equipment. Therefore, the load balancing schemes in data center should be resilient to network asymmetry. Prior fine-grained schemes such as RPS and Presto are prone to experience packet reordering problem under asymmetric topology since they split flows into small units which are spread across all parallel paths. The coarse-grained solutions such as ECMP and LetFlow effectively avoid packet reordering, but easily leading to under-utilization of multiple paths. To solve these problems, we propose a load balancing mechanism called AG, which adaptively adjusts switching granularity according to the asymmetric degree of multiple paths. AG increases switching granularity to alleviate packet reordering under large degrees of topology asymmetry, while reducing switching granularity to obtain high link utilization under small degrees of topology asymmetry. Moreover, we design a switch-based scheme which measures the difference of one-way delay of multiple paths to obtain accurate state of topology asymmetry with low overhead. AG is a practical switch-based solution without modification at end hosts. The experimental results of NS2 simulations and real implementation show that AG reduces the average and$99^{th}$flow completion time by up to 54% and 65% compared with the state-of-the-art load balancing schemes, respectively.

Nowadays, the distributed key-value stores have become the basic building block for large-scale cloud applications. In large-scale distributed key-value stores, many key-value access operations, which will be processed in parallel on different servers, are usually generated for a single end-user request. Accordingly, the completion time of an end-user request is determined by the last completed key-value access operation. Scheduling the order of serving key-value access operations can effectively reduce the completion times of end requests, thereby improving the user experience. However, existing scheduling algorithms hardly achieve consistent low latency due to the following challenges: the large overhead of cooperating clients and servers, the time-varying load and performance of servers, the traffic distribution can be either heavy-tailed or light-tailed and both the mean and the tail completion time are expected to be low. In this paper, we formalize the problem of scheduling key-value access operations and show it is NP-hard. Furthermore, we heuristically design the distributed adaptive scheduler (DAS), which distributively combines the largest remaining processing time last and the shortest remaining process time first algorithms. Theoretical analysis shows that DAS is adaptive to the time-varying traffic and server performance and can achieve consistent low mean and tail latency regardless of traffic distributions. Extensive simulations show that DAS reduces the mean request completion time by$17 \! \sim \! 50\%$with heavy-tailed traffic and$2 \! \sim 26 \! \%$with light-tailed traffic, while keeping the smallest tail completion time, compared to the default first come first served algorithm. Moreover, DAS outperforms the existing Rein-SBF algorithm under various scenarios.

In distributed key-value stores, multiple replica servers are always available for each key-value access operation when the eventual consistency model is employed. Accordingly, the completion times of the key-value access operations generated by an end-user request at different servers may be of great difference, especially when the replica servers are heterogeneous and have time-varying performance. Accordingly, the replica selection algorithm is crucial to cut the response time of end-user requests. The main challenge of making replica selection for each light-weighted key-value access operation is to timely know the status of replica serves. Recently, the adaptive replica selection algorithm C3 suggests guiding the replica selection with the piggybacked information of replica server in the returned “value”. Although C3 has good performance, the poor timeliness of feedback information makes a large performance gap between C3 and the ideal replica selection algorithm. To narrow this gap, the Accelerated Information Dissemination (AID) mechanism is proposed in this paper. Specifically, AID removes the bottleneck of information dissemination at the “slow” servers by letting both “client” and “replica server” store the records about the status of replica servers and both “key” and “value” piggyback multiple records. AID is implemented in Cassandra and evaluated by experiments and large scale simulations. The results show AID can significantly improve the timeliness of feedback information, especially when the number of nodes is large. Accordingly, AID helps C3 to greatly reduce the latency.

Recent methods of patient clinical outcome prediction focus on embedding the temporal time-series data by sequential data encoders without considering the dependency between the different variables and the static demographics data. To solve this problem and achieve better patient outcome prediction, we propose an attention-based memory fusion (AMF) network with Gated Recurrent Unit (GRU) (called GRU-AMFN) to model the dependency between the different time-series and static demographic data and extract effective personalized representation about the patient’s clinical health status. We evaluate our proposed GRU-AMFN method on eICU, a publicly available dataset, to validate its effectiveness for the in-hospital mortality prediction task. Experimental results demonstrate that our proposed method outperforms several state-of-the-art models for the in-hospital mortality prediction task. Ablation studies show the effectiveness of the proposed attention-based memory fusion module and the adaptive fusion module. Besides, our proposed method finds several static demographic and time-series features that are important for mortality prediction.

Biomedical named entity recognition (BioNER) is to extract entities, such as genes and proteins, from biomedical texts, where there is often a lack of high-quality training data. Recent work addresses this issue by multi-task learning with multiple datasets. However, these methods are usually over-parameterized, and some even suffer from negative transfer issues. To address above problems, we propose adversarial sparse sharing mechanism, which trains a sparsely shared encoder on multiple tasks with both task-agnostic and task-specific subnetworks. With adversarial training, we guide the task-agnostic subnetwork to learn shared task-invariant features and the task-specific subnetwork to learn task-dependent features. For a particular task, only the shared and its subnetwork are activated, which greatly reduces the number of parameters and avoids interference among tasks. Experimental results on 15 benchmark BioNER datasets show that our proposed method outperforms or is competitive with baseline methods with fewer parameters. Our code is released at: https://github.com/CSU-NLP-Group/ASNet

Protein-ligand interactions (PLIs) play important roles in cellular activities and drug discovery. Due to the technical difficulty and high cost of experimental methods, there is considerable interest in the development of computational approaches, such as protein-ligand docking, to decipher PLI patterns. One of the most important and difficult aspects of protein-ligand docking is recognizing near-native conformations from a set of decoys, but unfortunately, traditional scoring functions still suffer from limited accuracy. Therefore, new scoring methods are pressingly needed in methodological and/or practical implications. We present a new deep learning-based scoring function for ranking protein-ligand docking models based on Vision Transformer(ViT), named ViTRMSE. To recognize near-native conformations from a set of decoys, ViTRMSE voxelizes the protein-ligand interactional pocket into a 3D grid labeled by the occupancy contribution of atoms in different physicochemical classes. Benefiting from the Vision Transformer architecture, ViTRMSE can effectively capture the subtle differences between spatially and energetically favorable near-native conformations and unfavorable non-native decoys without needing extra information. ViTRMSE is extensively evaluated on diverse test sets including PDBbind2019 and CASF2016, and obtains significant improvements over existing methods in terms of RMSE, R and docking power.

Ki67 is a promising molecular biomarker for the diagnosis of lung adenocarcinoma. However, previous methods to determine Ki67 status often require tumor tissue sampling, which is invasive for patients. This study proposes a multi-view signature fusion network (MVSF), combining deep learning encoded (DLE) signatures, handcrafted radiomics (HCR) signatures, and clinical information to noninvasively predict Ki67 status. Multi-view signatures are combined through a tensor fusion network to obtain potentially high-dimensional signatures. Finally, a cooperative game theory-based approach is applied to quantitatively interpret the contribution of signatures to decision-making. The proposed MVSF is evaluated on a retrospectively collected dataset of 661 patients. Experimental results show that the MVSF achieves encouraging performance, with an area under the receiver operating characteristic curve of 0.80 and an accuracy of 0.78, outperforming several state-of-the-art Ki67 status prediction methods, which implies that our proposed method could provide potential support for Ki67 status prediction.

High-throughput biological and large-scale experiments for protein–protein interaction (PPI) identification provides valuable information about PPI networks but are time-consuming and limited in determining PPI between different species. We propose the integration of deep learning with feature fusion for PPI prediction, combining handcrafted features and protein sequence embedding for protein sequence representation. Our proposed method is evaluated on the Yeast full, Yeast core, Human, and eight independent datasets. The experimental results show that our method achieves 95.8% accuracy on the Yeast core, 99.2% accuracy on Human, and 100% accuracy on eight independent datasets. We also perform extensive comparisons with other existing outstanding methods and demonstrate the superior ability of the proposed method.

Resting-state functional magnetic resonance imaging (rs-fMRI) images have been widely used for diagnosis of schizophrenia. With rs-fMRI, most existing schizophrenia diagnostic methods have revealed schizophrenia’s functional abnormalities from the following three scales, i.e., regional neural activity alterations, functional connectivity abnormalities and brain network dysfunctions. However, many schizophrenia diagnosis methods do not consider the fusion of features from the three scales. In this study, we propose a schizophrenia diagnostic method based on multi-scale feature representation and ensemble learning. Firstly, features including the three scales (region, connectivity and network) are extracted from rs-fMRI images using the brainnetome atlas. For each scale, feature selection, i.e., least absolute shrinkage and selection operator, is applied to identify effective sub-features related to schizophrenia classification by a grid search. Then the selected sub-features of each scale are input to support vector machine with linear kernel to classify schizophrenia patients and healthy controls respectively. To further improve the schizophrenia diagnostic performance, an ensemble learning framework named E-RCN is proposed to average the probabilities obtained by the classifiers of each scale in decision level. By leave-one-out cross-validation on the center for biomedical research excellence dataset (COBRE), our proposed method achieves encouraging diagnosis performance, outperforming several state-of-the-art methods. In addition, ranked by the occurence frequency of each brain region within the leave-one-out cross-validation experiments, some brain regions related to schizophrenia, i.e., thalamus and middle temporal gyrus, and important elaborate subregions, i.e., Tha_L_8_8, MTG_L_4_4 and MTG_R_4_4, are found.

Identifying microbe-drug associations is important to systematically understand a drug’s mechanism of action in the therapeutic application. Since identifying microbe-drug associations is expensive and time-consuming via biological experiments, in this study, we propose a Pre-completion-based Label Propagation (PLP) method (called PLPMDA) to predict microbe-drug associations based on the multi-type similarities. To obtain richer information of drugs and microbes, we calculate drug chemical structure similarity, drug Anatomical Therapeutic Chemical (ATC) code similarity, microbe functional similarity, microbe sequence similarity and Gaussian Interaction Profile (GIP) kernel similarities of microbes and drugs, and then introduce a non-linear similarity fusion method. Comparing baseline methods, our advantage lies in performing an effective pre-completion step on the initial association matrix from the drug-related and microbe-related information and does not rely on the known drug-microbe associations, which can accelerate the design and discovery of the new drugs. The computational experiment results demonstrate that our proposed approach PLPMDA achieves significantly higher performance than the comparative methods in de novo and cross-validation experiments.

Receptor-binding is the first step of viral infection. Discovering potential virus-receptor interactions may give insight into potential strategies for treating viral infectious diseases. Most of computational methods for the virus-receptor interaction prediction are mainly based on sequence information. They neither makes effective use of structure information nor effectively handles with missing values of multiple similarities. In addition, the Link Prediction via linear optimization (LP) only uses contribution of neighbors of a node and ignores contribution of neighbors of another node on the network link. In this article, we present a virus-receptor interaction prediction method (MVLP) based on Multi-View learning and LP via contributions of all neighbors of two nodes on the network link. First, missing values of the receptor secondary structure similarity, the receptor conserved domain secondary structure similarity, the viral protein secondary structure similarity, the viral protein sequence similarity and the viral genome sequence similarity are updated by the gaussian radial basis function (GRB). To improve these similarities, we fuse updated and initial values of each similarity with multi-view learning, respectively. Next, three virus values and receptor similarities are integrated into the comprehensive virus and receptor similarity by the averaging method, respectively. Finally, LP based on contribution of neighbors of two nodes is presented for the virus-receptor interaction prediction. To evaluate the ability of MVLP, we compare MVLP with four related methods in 10 fold Cross-Validation (10CV). Computational results indicate that an average Area Under Curve (AUC) values of MVLP on viralReceptor sup and viralReceptor are 0.9427 and 0.9444, respectively, which are superior to other related methods. Furthermore, a case study also demonstrates the ability of MVLP in practice.

Urban flow imputation, which aims to infer the missing flows of some locations based on the available flows of surrounding areas, is critically important to various smart city related applications such as urban planning and public safety. Although many methods are proposed to impute time series data, they may not be feasible to be directly applied on urban flow data due to the following reasons. First, urban flows have the complex spatial and temporal correlations which are much harder to be captured compared with time series data. Second, the urban flow data can be random missing (i.e., missing randomly in terms of times and locations) or block missing (i.e., missing for all locations in a particular time slot). Thus it is difficult for existing methods to work well on both scenarios. In this paper, we for the first time study the urban flow imputation problem and propose a generative-free Attention-based Spatial-Temporal Combine and Mix Completion Network model (AST-CMCN for short) to effectively address it. Specifically, AST-CMCN consists of a Spatial and Temporal Completion Network (SATCNet for short) and a Spatial-Temporal Mix Completion Network (STMCNet for short). SATCNet is composed of stacked GRUAtt modules to capture the geographical and temporal correlations of the urban flows, separately. STMCNet is designed to capture the complex spatial-temporal associations jointly between historical urban flows and current data. A Message Passing module is also proposed to capture new spatial-temporal patterns that never appear in the historical data. Extensive experiments on two large real-world datasets validate the effectiveness and efficiency of our method compared with the state-of-the-art baselines.

In the parameter-server-based distributed deep learning system, the workers simultaneously communicate with the parameter server to refine model parameters, easily resulting in severe network contention. To solve this problem, Asynchronous Parallel (ASP) strategy enables each worker to update the parameter independently without synchronization. However, due to the inconsistency of parameters among workers, ASP experiences accuracy loss and slow convergence. In this paper, we propose Hybrid Synchronous Parallelism (HSP), which mitigates the communication contention without excessive degradation of convergence speed. Specifically, the parameter server sequentially pulls gradients from workers to eliminate network congestion and synchronizes all up-to-date parameters after each iteration. Meanwhile, HSP cautiously lets idle workers to compute with out-of-date weights to maximize the utilizations of computing resources. We provide theoretical analysis of convergence efficiency and implement HSP on popular deep learning (DL) framework. The test results show that HSP improves the convergence speedup of three classical deep learning models by up to 67%.

The k-means problem is an extensively studied unsupervised learning problem with various applications in decision making and data mining. In this paper, we propose a fast and practical local search algorithm for the k-means problem. Our method reduces the search space of swap pairs from O(nk) to O(k^2), and applies random mutations to find potentially better solutions when local search falls into poor local optimum. With the assumption of data distribution that each optimal cluster has "average" size of \Omega(n/k), which is common in many datasets and k-means benchmarks, we prove that our proposed algorithm gives a (100+\epsilon)-approximate solution in expectation. Empirical experiments show that our algorithm achieves better performance compared to existing state-of-the-art local search methods on k-means benchmarks and large datasets.

Copa is a delay-based congestion control algorithm proposed in NSDI recently. It can achieve consistent high performance under various network environments and has already been deployed in Facebook. In this paper, we theoretically analyze Copa and reveal its large queuing delay and poor fairness issue under certain conditions. The root cause is that Copa fails to clear the bottleneck buffer occupancy periodically as expected. Accordingly, Copa may get a wrong base RTT estimation and enter its competitive mode by mistake, leading to large delay and unfairness. To address these issues, we propose Copa+, which enhances Copa with a parameter adaptation mechanism and an optimized competitive mode entrance criterion. Designed based on our theoretical analysis, Copa+ can adaptively clear the bottleneck buffer occupancy for correct estimation of base RTT. Consequently, Copa+ inherits the advantages of Copa but achieves lower queuing delay and better fairness under different environments, as confirmed by the real-world experiments and simulations. Specifically, Copa+ has the highest throughput similar to Copa but 11.9% lower queuing delay over different Internet links among different cloud nodes, and achieves 39.4% lower queuing delay and 8.9% higher throughput compared to Sprout over emulated cellular links.

A fundamental challenge in deep metric learning is the generalization capability of the feature embedding network model since the embedding network learned on training classes need to be evaluated on new test classes. To address this challenge, in this paper, we introduce a new method called coded residual transform (CRT) for deep metric learning to significantly improve its generalization capability. Specifically, we learn a set of diversified prototype features, project the feature map onto each prototype, and then encode its features using their projection residuals weighted by their correlation coefficients with each prototype. The proposed CRT method has the following two unique characteristics. First, it represents and encodes the feature map from a set of complimentary perspectives based on projections onto diversified prototypes. Second, unlike existing transformer-based feature representation approaches which encode the original values of features based on global correlation analysis, the proposed coded residual transform encodes the relative differences between the original features and their projected prototypes. Embedding space density and spectral decay analysis show that this multi perspective projection onto diversified prototypes and coded residual representation are able to achieve significantly improved generalization capability in metric learning. Finally, to further enhance the generalization performance, we propose to enforce the consistency on their feature similarity matrices between coded residual transforms with different sizes of projection prototypes and embedding dimensions. Our extensive experimental results and ablation studies demonstrate that the proposed CRT method outperform the state-of-the-art deep metric learning methods by large margins and improving upon the current best method by up to 4.28% on the CUB dataset.

Long-read sequencing technology enables significant progress in de novo genome assembly. However, the high error rate and the wide error distribution of raw reads result in a large number of errors in the assembly. Polishing is a procedure to fix errors in the draft assembly and improve the reliability of genomic analysis. However, existing methods treat all the regions of the assembly equally while there are fundamental differences between the error distributions of these regions. How to achieve very high accuracy in genome assembly is still a challenging problem. Motivated by the uneven errors in different regions of the assembly, we propose a novel polishing workflow named BlockPolish. In this method, we divide contigs into blocks with low complexity and high complexity according to statistics of aligned nucleotide bases. Multiple sequence alignment is applied to realign raw reads in complex blocks and optimize the alignment result. Due to the different distributions of error rates in trivial and complex blocks, two multitask bidirectional Long short-term memory (LSTM) networks are proposed to predict the consensus sequences. In the whole-genome assemblies of NA12878 assembled by Wtdbg2 and Flye using Nanopore data, BlockPolish has a higher polishing accuracy than other state-of-the-arts including Racon, Medaka and MarginPolish & HELEN. In all assemblies, errors are predominantly indels and BlockPolish has a good performance in correcting them. In addition to the Nanopore assemblies, we further demonstrate that BlockPolish can also reduce the errors in the PacBio assemblies. The source code of BlockPolish is freely available on Github (https://github.com/huangnengCSU/BlockPolish).

Alzheimer's disease (AD) has a strong genetic predisposition. However, its risk genes remain incompletely identified. We developed an Alzheimer's brain gene network-based approach to predict AD-associated genes by leveraging the functional pattern of known AD-associated genes. Our constructed network outperformed existing networks in predicting AD genes. We then systematically validated the predictions using independent genetic, transcriptomic, proteomic data, neuropathological and clinical data. First, top-ranked genes were enriched in AD-associated pathways. Second, using external gene expression data from the Mount Sinai Brain Bank study, we found that the top-ranked genes were significantly associated with neuropathological and clinical traits, including the Consortium to Establish a Registry for Alzheimer's Disease score, Braak stage score and clinical dementia rating. The analysis of Alzheimer's brain single-cell RNA-seq data revealed cell-type-specific association of predicted genes with early pathology of AD. Third, by interrogating proteomic data in the Religious Orders Study and Memory and Aging Project and Baltimore Longitudinal Study of Aging studies, we observed a significant association of protein expression level with cognitive function and AD clinical severity. The network, method and predictions could become a valuable resource to advance the identification of risk genes for AD.

Investigating differentially methylated regions (DMRs) presented in different tissues or cell types can help to reveal the mechanisms behind the tissue-specific gene expression. The identified tissue-/disease-specific DMRs also can be used as feature markers for spotting the tissues-of-origins of cell-free DNA (cfDNA) in noninvasive diagnosis. In recent years, many methods have been proposed to detect DMRs. However, due to the lack of benchmark DMRs, it is difficult for researchers to choose proper methods and select desirable DMR sets for downstream studies. The application of DMRs, used as feature markers, can be benefited by the longer length of DMRs containing more CpG sites when a threshold is given for the methylation differences of DMRs. According to this, two metrics ($Qn$ and $Ql$), in which the CpG numbers and lengths of DMRs with different methylation differences are weighted differently, are proposed in this paper to evaluate the DMR sets predicted by different methods on BS-seq data. DMR sets predicted by eight methods on both simulated datasets and real BS-seq datasets are evaluated by the proposed metrics, the benchmark-based metrics, and the enrichment analysis of biological data, including genomic features, transcription factors and histones. The rank correlation analysis shows that the $Qn$ and $Ql$ are highly correlated to the benchmark metrics for simulated datasets and the biological data enrichment analysis for real BS-seq data. Therefore, with no need for additional biological data, the proposed metrics can help researchers selecting a more suitable DMR set on a certain BS-seq dataset.

Determining drug indications is a critical part of the drug development process. However, traditional drug discovery is expensive and time-consuming. Drug repositioning aims to find potential indications for existing drugs, which is considered as an important alternative to the traditional drug discovery. In this article, we propose a multi-view learning with matrix completion (MLMC) method to predict the potential associations between drugs and diseases. Specifically, MLMC first learns the comprehensive similarity matrices from five drug similarity matrices and two disease similarity matrices based on the multi-view learning (ML) with Laplacian graph regularization, and updates the drug-disease association matrix simultaneously. Then, we introduce matrix completion (MC) to add some positive entries in original association matrix based on low-rank structure, and re-execute the multi-view learning algorithm for association prediction. At last, the prediction results of the above two operations are integrated as the final output. Evaluated by 10-fold cross-validation and de novo tests, MLMC achieves higher prediction accuracy than the current state-of-the-art methods. Moreover, case studies confirm the ability of our method in novel drug-disease association discovery. The codes of MLMC are available at https://github.com/BioinformaticsCSU/MLMC. Contact: [email protected].

Proteins maintain the functional order of cell in life by interacting with other proteins. Determination of protein complex structural information gives biological insights for the research of diseases and drugs. Recently, a breakthrough has been made in protein monomer structure prediction. However, due to the limited number of the known protein structure and homologous sequences of complexes, the prediction of residue-residue contacts on hetero-dimer interfaces is still a challenge. In this study, we have developed a deep learning framework for inferring inter-protein residue contacts from sequential information, called HDIContact. We utilized transfer learning strategy to produce Multiple Sequence Alignment (MSA) two-dimensional (2D) embedding based on patterns of concatenated MSA, which could reduce the influence of noise on MSA caused by mismatched sequences or less homology. For MSA 2D embedding, HDIContact took advantage of Bi-directional Long Short-Term Memory (BiLSTM) with two-channel to capture 2D context of residue pairs. Our comprehensive assessment on the Escherichia coli (E. coli) test dataset showed that HDIContact outperformed other state-of-the-art methods, with top precision of 65.96%, the Area Under the Receiver Operating Characteristic curve (AUROC) of 83.08% and the Area Under the Precision Recall curve (AUPR) of 25.02%. In addition, we analyzed the potential of HDIContact for human-virus protein-protein complexes, by achieving top five precision of 80% on O75475-P04584 related to Human Immunodeficiency Virus. All experiments indicated that our method was a valuable technical tool for predicting inter-protein residue contacts, which would be helpful for understanding protein-protein interaction mechanisms.

The side effects of drugs present growing concern attention in the healthcare system. Accurately identifying the side effects of drugs is very important for drug development and risk assessment. Some computational models have been developed to predict the potential side effects of drugs and provided satisfactory performance. However, most existing methods can only predict whether side effects will occur and cannot determine the frequency of side effects. Although a few existing methods can predict the frequency of drug side effects, they strongly depend on the known drug-side effect relationships. Therefore, they cannot be applied to new drugs without known side effect frequency information. In this paper, we develop a novel similarity-based deep learning method, named SDPred, for determining the frequencies of drug side effects. Compared with the existing state-of-the-art models, SDPred integrates rich features and can be applied to predict the side effect frequencies of new drugs without any known drug-side effect association or frequency information. To our knowledge, this is the first work that can predict the side effect frequencies of new drugs in the population. The comparison results indicate that SDPred is much superior to all previously reported models. In addition, some case studies also demonstrate the effectiveness of our proposed method in practical applications. The SDPred software and data are freely available at https://github.com/zhc940702/SDPred, https://zenodo.org/record/5112573 and https://hub.docker.com/r/zhc940702/sdpred.

As a frontier field of individualized therapy, microRNA (miRNA) pharmacogenomics facilitates the understanding of different individual responses to certain drugs and provides a reasonable reference for clinical treatment. However, the known drug resistance-associated miRNAs are not yet sufficient to support precision medicine. Although existing methods are effective, they all focus on modelling miRNA-drug resistance interaction graphs, making their performance bounded by the interaction density. In this study, we propose a framework for miRNA-drug resistance prediction through efficient neural architecture search and graph isomorphism networks (NASMDR). NASMDR uses attribute information instead of the commonly used interactive graph information. In the cross-validation experiment, the proposed framework can achieve an AUC of 0.9468 on the ncDR dataset, which is 2.29% higher than the state-of-the-art method. In addition, we propose a novel sequence characterization approach, k-mer Sparse Nonnegative Matrix Factorization (KSNMF). The results show that NASMDR provides novel insights for integrating efficient neural architecture search and graph isomorphic networks into a unified framework to predict drug resistance-related miRNAs. The codes for NASMDR are available at https://github.com/kaizheng-academic/NASMDR.

MOTIVATION: Identifying drug-target interactions is a crucial step for drug discovery and design. Traditional biochemical experiments are credible to accurately validate drug-target interactions. However, they are also extremely laborious, time-consuming and expensive. With the collection of more validated biomedical data and the advancement of computing technology, the computational methods based on chemogenomics gradually attract more attention, which guide the experimental verifications. RESULTS: In this study, we propose an end-to-end deep learning-based method named IIFDTI to predict drug-target interactions (DTIs) based on independent features of drug-target pairs and interactive features of their substructures. First, the interactive features of substructures between drugs and targets are extracted by the bidirectional encoder-decoder architecture. The independent features of drugs and targets are extracted by the graph neural networks and convolutional neural networks, respectively. Then, all extracted features are fused and inputted into fully connected dense layers in downstream tasks for predicting DTIs. IIFDTI takes into account the independent features of drugs/targets and simulates the interactive features of the substructures from the biological perspective. Multiple experiments show that IIFDTI outperforms the state-of-the-art methods in terms of the area under the receiver operating characteristics curve (AUC), the area under the precision-recall curve (AUPR), precision, and recall on benchmark datasets. In addition, the mapped visualizations of attention weights indicate that IIFDTI has learned the biological knowledge insights, and two case studies illustrate the capabilities of IIFDTI in practical applications. AVAILABILITY AND IMPLEMENTATION: The data and codes underlying this article are available in Github at https://github.com/czjczj/IIFDTI. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: Protein-protein interactions (PPI) play important roles in cellular activities. Due to the technical difficulty and high cost of experimental methods, there are considerable interests towards the development of computational approaches, such as protein docking, to decipher PPI patterns. One of the important and difficult aspects in protein docking is recognizing near-native conformations from a set of decoys, but unfortunately, traditional scoring functions still suffer from limited accuracy. Therefore, new scoring methods are pressingly needed in methodological and/or practical implications. RESULTS: We present a new deep learning-based scoring method for ranking protein-protein docking models based on a 3D RepVGG network, named TRScore. To recognize near-native conformations from a set of decoys, TRScore voxelizes the protein-protein interface into a 3D grid labeled by the number of atoms in different physicochemical classes. Benefiting from the deep convolutional RepVGG architecture, TRScore can effectively capture the subtle differences between energetically favorable near-native models and unfavorable non-native decoys without needing extra information. TRScore was extensively evaluated on diverse test sets including protein-protein docking benchmark 5.0 update set, DockGround decoy set, as well as realistic CAPRI decoy set and overall obtained a significant improvement over existing methods in cross-validation and independent evaluations. AVAILABILITY AND IMPLEMENTATION: Codes available at: https://github.com/BioinformaticsCSU/TRScore.

MOTIVATION: Alzheimer's disease (AD) is a complex brain disorder with risk genes incompletely identified. The candidate genes are dominantly obtained by computational approaches. In order to obtain biological insights of candidate genes or screen genes for experimental testing, it is essential to assess their relevance to AD. A platform that integrates different types of omics data and approaches would facilitate the analysis of candidate genes and is in great need. RESULTS: We report AlzCode, a platform for multiview analysis of genes related to AD. First, this platform integrates a rich collection of functional genomic data, including expression data of AD samples (gene expression, single-cell RNA-seq data and protein expression), AD-specific biological networks (co-expression networks and functional gene networks), neuropathological and clinical traits (CERAD score, Braak staging score, Clinical Dementia Rating, cognitive function and clinical severity) and general data such as protein-protein interaction, regulatory networks, sequence similarity and miRNA-target interactions. These data provide basis for analyzing genes from different views. Second, the platform integrates multiple approaches designed for the various types of data. We implement functions to analyze both individual genes and gene sets. We also compare AlzCode with two existing platforms for AD analysis, which are Agora and AD Atlas. We pinpoint the features of each platform and highlight their differences. This platform would be valuable to the understanding of AD genetics and pathological mechanisms. AVAILABILITY AND IMPLEMENTATION: AlzCode is freely available at: http://www.alzcode.xyz. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: Due to cancer heterogeneity, the therapeutic effect may not be the same when a cohort of patients of the same cancer type receive the same treatment. The anticancer drug response prediction may help develop personalized therapy regimens to increase survival and reduce patients' expenses. Recently, graph neural network-based methods have aroused widespread interest and achieved impressive results on the drug response prediction task. However, most of them apply graph convolution to process cell line-drug bipartite graphs while ignoring the intrinsic differences between cell lines and drug nodes. Moreover, most of these methods aggregate node-wise neighbor features but fail to consider the element-wise interaction between cell lines and drugs. RESULTS: This work proposes a neighborhood interaction (NI)-based heterogeneous graph convolution network method, namely NIHGCN, for anticancer drug response prediction in an end-to-end way. Firstly, it constructs a heterogeneous network consisting of drugs, cell lines and the known drug response information. Cell line gene expression and drug molecular fingerprints are linearly transformed and input as node attributes into an interaction model. The interaction module consists of a parallel graph convolution network layer and a NI layer, which aggregates node-level features from their neighbors through graph convolution operation and considers the element-level of interactions with their neighbors in the NI layer. Finally, the drug response predictions are made by calculating the linear correlation coefficients of feature representations of cell lines and drugs. We have conducted extensive experiments to assess the effectiveness of our model on Cancer Drug Sensitivity Data (GDSC) and Cancer Cell Line Encyclopedia (CCLE) datasets. It has achieved the best performance compared with the state-of-the-art algorithms, especially in predicting drug responses for new cell lines, new drugs and targeted drugs. Furthermore, our model that was well trained on the GDSC dataset can be successfully applied to predict samples of PDX and TCGA, which verified the transferability of our model from cell line in vitro to the datasets in vivo. AVAILABILITY AND IMPLEMENTATION: The source code can be obtained from https://github.com/weiba/NIHGCN. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: Many studies have shown that microRNAs (miRNAs) play a key role in human diseases. Meanwhile, traditional experimental methods for miRNA-disease association identification are extremely costly, time-consuming and challenging. Therefore, many computational methods have been developed to predict potential associations between miRNAs and diseases. However, those methods mainly predict the existence of miRNA-disease associations, and they cannot predict the deep-level miRNA-disease association types. RESULTS: In this study, we propose a new end-to-end deep learning method (called PDMDA) to predict deep-level miRNA-disease associations with graph neural networks (GNNs) and miRNA sequence features. Based on the sequence and structural features of miRNAs, PDMDA extracts the miRNA feature representations by a fully connected network (FCN). The disease feature representations are extracted from the disease-gene network and gene-gene interaction network by GNN model. Finally, a multilayer with three fully connected layers and a softmax layer is designed to predict the final miRNA-disease association scores based on the concatenated feature representations of miRNAs and diseases. Note that PDMDA does not take the miRNA-disease association matrix as input to compute the Gaussian interaction profile similarity. We conduct three experiments based on six association type samples (including circulations, epigenetics, target, genetics, known association of which their types are unknown and unknown association samples). We conduct fivefold cross-validation validation to assess the prediction performance of PDMDA. The area under the receiver operating characteristic curve scores is used as metric. The experiment results show that PDMDA can accurately predict the deep-level miRNA-disease associations. AVAILABILITY AND IMPLEMENTATION: Data and source codes are available at https://github.com/27167199/PDMDA.

The third-generation sequencing technology has advanced genome analysis with long-read length, but the reads need error correction due to the high error rate. Error correction is a time-consuming process especially when the sequencing coverage is high. Generally, for a pair of overlapping reads A and B, the existing error correction methods perform a base-level alignment from B to A when correcting the read A. And another base-level alignment from A to B is performed when correcting the read B. However, based on our observation, the base-level alignment information can be reused. In this article, we present a fast error correction tool Fec, using two-rounds overlapping and caching. Fec can be used independently or as an error correction step in an assembly pipeline. In the first round, Fec uses a large window size (20) to quickly find enough overlaps to correct most of the reads. In the second round, a small window size (5) is used to find more overlaps for the reads with insufficient overlaps in the first round. When performing base-level alignment, Fec searches the cache first. If the alignment exists in the cache, Fec takes this alignment out and deduces the second alignment from it. Otherwise, Fec performs base-level alignment and stores the alignment in the cache. We test Fec on nine datasets, and the results show that Fec has 1.24-38.56 times speed-up compared to MECAT, CANU and MINICNS on five PacBio datasets and 1.16-27.8 times speed-up compared to NECAT and CANU on four nanopore datasets. AVAILABILITY AND IMPLEMENTATION: Fec is available at https://github.com/zhangjuncsu/Fec. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: Identifying drug-target interactions (DTIs) is a crucial step in drug repurposing and drug discovery. Accurately identifying DTIs in silico can significantly shorten development time and reduce costs. Recently, many sequence-based methods are proposed for DTI prediction and improve performance by introducing the attention mechanism. However, these methods only model single non-covalent inter-molecular interactions among drugs and proteins and ignore the complex interaction between atoms and amino acids. RESULTS: In this article, we propose an end-to-end bio-inspired model based on the convolutional neural network (CNN) and attention mechanism, named HyperAttentionDTI, for predicting DTIs. We use deep CNNs to learn the feature matrices of drugs and proteins. To model complex non-covalent inter-molecular interactions among atoms and amino acids, we utilize the attention mechanism on the feature matrices and assign an attention vector to each atom or amino acid. We evaluate HpyerAttentionDTI on three benchmark datasets and the results show that our model achieves significantly improved performance compared with the state-of-the-art baselines. Moreover, a case study on the human Gamma-aminobutyric acid receptors confirm that our model can be used as a powerful tool to predict DTIs. AVAILABILITY AND IMPLEMENTATION: The codes of our model are available at https://github.com/zhaoqichang/HpyerAttentionDTI and https://zenodo.org/record/5039589. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Predicting the future risk of cardiovascular diseases from the historical Electronic Health Records (EHRs) is a significant research task in personalized healthcare fields. In recent years, many deep neural network-based methods have emerged, which model patient disease progression by capturing the temporal patterns in sequential visit data. However, most existing methods that integrate different types of clinical data often do not fully consider the impact of the patient's age and the irregular time interval between consecutive medical records on the patient's disease development. To address these challenges, we propose a Time-Aware Multi-type Data fUsion Representation learning framework (TAMDUR) for cardiovascular diseases (CVDs) risk prediction. In this framework, we design a time-aware decay function, which is based on the patient's age and the elapsed time between visits, to model the disease progression pattern. Then, a parallel combination of bidirectional long short-term memory (Bi_LSTM) network and convolutional neural network (CNN) is constructed to respectively learn the temporal and non-temporal features from various types of clinical data. Finally, a multi-type data fusion representation layer based on self-attention is utilized to integrate various features and their correlations to obtain the final patient representation. We evaluate our model on a real medical dataset, and the experimental results demonstrate that TAMDUR outperforms the state-of-the-art approaches.

The accurate prediction of glioma grade before surgery is essential for treatment planning and prognosis. Since the gold standard (i.e., biopsy)for grading gliomas is both highly invasive and expensive, and there is a need for a noninvasive and accurate method. In this study, we proposed a novel radiomics-based pipeline by incorporating the intratumoral and peritumoral features extracted from preoperative mpMRI scans to accurately and noninvasively predict glioma grade. To address the unclear peritumoral boundary, we designed an algorithm to capture the peritumoral region with a specified radius. The mpMRI scans of 285 patients derived from a multi-institutional study were adopted. A total of 2153 radiomic features were calculated separately from intratumoral volumes (ITVs)and peritumoral volumes (PTVs)on mpMRI scans, and then refined using LASSO and mRMR feature ranking methods. The top-ranking radiomic features were entered into the classifiers to build radiomic signatures for predicting glioma grade. The prediction performance was evaluated with five-fold cross-validation on a patient-level split. The radiomic signatures utilizing the features of ITV and PTV both show a high accuracy in predicting glioma grade, with AUCs reaching 0.968. By incorporating the features of ITV and PTV, the AUC of IPTV radiomic signature can be increased to 0.975, which outperforms the state-of-the-art methods. Additionally, our proposed method was further demonstrated to have strong generalization performance in an external validation dataset with 65 patients. The source code of our implementation is made publicly available at https://github.com/chengjianhong/glioma_grading.git.

Identifying drug-target interactions (DTIs) is an important step in the process of new drug discovery and drug repositioning. Accurate predictions for DTIs can improve the efficiency in the drug discovery and development. Although rapid advances in deep learning technologies have generated various computational methods, it is still appealing to further investigate how to design efficient networks for predicting DTIs. In this study, we propose an end-to-end deep learning method (called MHSADTI) to predict DTIs based on the graph attention network and multi-head self-attention mechanism. First, the characteristics of drugs and proteins are extracted by the graph attention network and multi-head self-attention mechanism, respectively. Then, the attention scores are used to consider which amino acid subsequence in a protein is more important for the drug to predict its interactions. Finally, we predict DTIs by a fully connected layer after obtaining the feature vectors of drugs and proteins. MHSADTI takes advantage of self-attention mechanism for obtaining long-dependent contextual relationship in amino acid sequences and predicting DTI interpretability. More effective molecular characteristics are also obtained by the attention mechanism in graph attention networks. Multiple cross validation experiments are adopted to assess the performance of our MHSADTI. The experiments on four datasets, human, C.elegans, DUD-E and DrugBank show our method outperforms the state-of-the-art methods in terms of AUC, Precision, Recall, AUPR and F1-score. In addition, the case studies further demonstrate that our method can provide effective visualizations to interpret the prediction results from biological insights.

Accurate and rapid diagnosis of coronavirus disease 2019 (COVID-19) from chest CT scans is of great importance and urgency during the worldwide outbreak. However, radiologists have to distinguish COVID-19 pneumonia from other pneumonia in a large number of CT scans, which is tedious and inefficient. Thus, it is urgently and clinically needed to develop an efficient and accurate diagnostic tool to help radiologists to fulfill the difficult task. In this study, we proposed a deep supervised autoencoder (DSAE) framework to automatically identify COVID-19 using multi-view features extracted from CT images. To fully explore features characterizing CT images from different frequency domains, DSAE was proposed to learn the latent representation by multi-task learning. The proposal was designed to both encode valuable information from different frequency features and construct a compact class structure for separability. To achieve this, we designed a multi-task loss function, which consists of a supervised loss and a reconstruction loss. Our proposed method was evaluated on a newly collected dataset of 787 subjects including COVID-19 pneumonia patients, other pneumonia patients, and normal subjects without abnormal CT findings. Extensive experimental results demonstrated that our proposed method achieved encouraging diagnostic performance and may have potential clinical application for the diagnosis of COVID-19.

The Coronavirus Disease 2019 (COVID-19) has become an international public health emergency, posing a serious threat to human health and safety around the world. The 2019-nCoV coronavirus spike protein was confirmed to be highly susceptible to various mutations, which can trigger apparent changes of virus transmission capacity and the pathogenic mechanism. In this article, the binding interface was obtained by analyzing the interaction modes between 2019-nCoV coronavirus and the human ACE2. Based on the "SIFT server" and the "bubble" identification mechanism, 9 amino acid sites were selected as potential mutation-sites from the 2019-nCoV-S1-ACE2 binding interface. Subsequently, a total number of 171 mutant systems for 9 mutation-sites were optimized for binding-pattern comparsion analysis, and 14 mutations that may improve the binding capacity of 2019-nCoV-S1 to ACE2 were selected. The Molecular Dynamic Simulations were conducted to calculate the binding free energies of all the 14 mutant systems. Finally, we found that most of the 14 mutations on the 2019-nCoV-S1 protein could enhance the binding ability between 2019-nCoV coronavirus and human ACE2. Among which, the binding capacities for G446R, Y449R and F486Y mutations could be increased by 20 percent, and that for S494R mutant increased even by 38.98 percent. We hope this research could provide significant help for the future epidemic detection, drug and vaccine development.

Gene regulatory networks (GRNs)are involved in various biological processes, such as cell cycle, differentiation and apoptosis. The existing large amount of expression data, especially the time-series expression data, provide a chance to infer GRNs by computational methods. These data can reveal the dynamics of gene expression and imply the regulatory relationships among genes. However, identify the indirect regulatory links is still a big challenge as most studies treat time points as independent observations, while ignoring the influences of time delays. In this study, we propose a GRN inference method based on information-theory measure, called NIMCE. NIMCE incorporates the transfer entropy to measure the regulatory links between each pair of genes, then applies the causation entropy to filter indirect relationships. In addition, NIMCE applies multi time delays to identify indirect regulatory relationships from candidate genes. Experiments on simulated and colorectal cancer data show NIMCE outperforms than other competing methods. All data and codes used in this study are publicly available at https://github.com/CSUBioGroup/NIMCE.

In recent years, cancer patients survival prediction holds important significance for worldwide health problems, and has gained many researchers attention in medical information communities. Cancer patients survival prediction can be seen the classification work which is a meaningful and challenging task. Nevertheless, research in this field is still limited. In this work, we design a novel Multimodal Graph Neural Network (MGNN)framework for predicting cancer survival, which explores the features of real-world multimodal data such as gene expression, copy number alteration and clinical data in a unified framework. Specifically, we first construct the bipartite graphs between patients and multimodal data to explore the inherent relation. Subsequently, the embedding of each patient on different bipartite graphs is obtained with graph neural network. Finally, a multimodal fusion neural layer is proposed to fuse the medical features from different modality data. Comprehensive experiments have been conducted on real-world datasets, which demonstrate the superiority of our modal with significant improvements against state-of-the-arts. Furthermore, the proposed MGNN is validated to be more robust on other four cancer datasets.

Highly portable Oxford Nanopore sequencer producing long reads in real-time at low cost has made many breakthroughs in genomics studies. However, a major limitation of nanopore sequencing is its high errors when deciphering DNA sequences from noisy and complex raw data. In this paper, we developed an end-to-end basecaller, SACall, based on convolution layers, transformer self-attention layers and a CTC decoder. In SACall, the convolution layers are used to downsample the signals and capture the local patterns. To achieve the contextual relevance of signals, self-attention layers are adopted to calculate the similarity of the signals at any two positions in the raw signal sequence. Finally, the CTC decoder generates the DNA sequence by a beam search algorithm. We use a benchmark consisting of nine isolated genomes to test the quality of different basecallers including SACall, Albacore, and Guppy. The performances of basecallers are evaluated from the perspective of read accuracy, assembly quality, and consensus accuracy. Among most of the genomes in the test benchmark, the reads basecalled by SACall have fewer errors than the reads basecalled by other basecallers. When assembling the basecalled reads of each genome, the assembly from SACall basecalled reads achieves a higher assembly identity. In addition, there are fewer errors in the polished assembly from reads basecalled by SACall compared to those basecalled by Albacore and Guppy. In general, SACall outperforms the Nanopore official basecallers Albacore and Guppy in the benchmark. Moreover, SACall is an open-source and freely available basecaller, which gives a chance for researchers to train their own basecalling models on specific data and basecall Nanopore reads.

Accumulating evidences have shown that circRNA plays an important role in human diseases. It can be used as potential biomarker for diagnose and treatment of disease. Although some computational methods have been proposed to predict circRNA-disease associations, the performance still need to be improved. In this paper, we propose a new computational model based on Improved Graph convolutional network and Negative Sampling to predict CircRNA-Disease Associations. In our method, it constructs the heterogeneous network based on known circRNA-disease associations. Then, an improved graph convolutional network is designed to obtain the feature vectors of circRNA and disease. Further, the multi-layer perceptron is employed to predict circRNA-disease associations based on the feature vectors of circRNA and disease. In addition, the negative sampling method is employed to reduce the effect of the noise samples, which selects negative samples based on circRNA's expression profile similarity and Gaussian Interaction Profile kernel similarity. The 5-fold cross validation is utilized to evaluate the performance of the method. The results show that IGNSCDA outperforms than other state-of-the-art methods in the prediction performance. Moreover, the case study shows that IGNSCDA is an effective tool for predicting potential circRNA-disease associations.

It has been proved that long noncoding RNA (lncRNA) plays critical roles in many human diseases. Therefore, inferring associations between lncRNAs and diseases can contribute to disease diagnosis, prognosis and treatment. To overcome the limitation of traditional experimental methods such as expensive and time-consuming, several computational methods have been proposed to predict lncRNA-disease associations by fusing different biological data. However, the prediction performance of lncRNA-disease associations identification needs to be improved. In this study, we propose a computational model (named LDICDL) to identify lncRNA-disease associations based on collaborative deep learning. It uses an automatic encoder to denoise multiple lncRNA feature information and multiple disease feature information, respectively. Then, the matrix decomposition algorithm is employed to predict the potential lncRNA-disease associations. In addition, to overcome the limitation of matrix decomposition, the hybrid model is developed to predict associations between new lncRNA (or disease) and diseases (or lncRNA). The ten-fold cross validation and de novo test are applied to evaluate the performance of method. The experimental results show LDICDL outperforms than other state-of-the-art methods in prediction performance.

BACKGROUND: In medicine, karyotyping chromosomes is important for medical diagnostics, drug development, and biomedical research. Unfortunately, chromosome karyotyping is usually done by skilled cytologists manually, which requires experience, domain expertise, and considerable manual efforts. Therefore, automating the karyotyping process is a significant and meaningful task. METHOD: This paper focuses on chromosome classification because it is critical for chromosome karyotyping. In recent years, deep learning-based methods are the most promising methods for solving the tasks of chromosome classification. Although the deep learning-based Inception architecture has yielded state-of-the-art performance in the 2015 ILSVRC challenge, it has not been used in chromosome classification tasks so far. Therefore, we develop an automatic chromosome classification approach named CIR-Net based on Inception-ResNet which is an optimized version of Inception. However, the classification performance of origin Inception-ResNet on the insufficient chromosome dataset still has a lot of capacity for improvement. Further, we propose a simple but effective augmentation method called CDA for improving the performance of CIR-Net. RESULTS: The experimental results show that our proposed method achieves 95.98 percent classification accuracy on the clinical G-band chromosome dataset whose training dataset is insufficient. Moreover, the proposed augmentation method CDA improves more than 8.5 percent (from 87.46 to 95.98 percent) classification accuracy comparing to other methods. In this paper, the experimental results demonstrate that our proposed method is recent the most effective solution for solving clinical chromosome classification problems in chromosome auto-karyotyping on the condition of the insufficient training dataset. Code and Dataset are available at https://github.com/CloudDataLab/CIR-Net.

Magnetic resonance imagings (MRIs) are providing increased access to neuropsychiatric disorders that can be made available for advanced data analysis. However, the single type of data limits the ability of psychiatrists to distinguish the subclasses of this disease. In this paper, we propose an ensemble hybrid features selection method for the neuropsychiatric disorder classification. The method consists of a 3D DenseNet and a XGBoost, which are used to select the image features from structural MRI images and the phenotypic feature from phenotypic records, respectively. The hybrid feature is composed of image features and phenotypic features. The proposed method is validated in the Consortium for Neuropsychiatric Phenomics (CNP) dataset, where samples are classified into one of the four classes (healthy controls (HC), attention deficit hyperactivity disorder (ADHD), bipolar disorder (BD), and schizophrenia (SD)). Experimental results show that the hybrid feature can improve the performance of classification methods. The best accuracy of binary and multi-class classification can reach 91.22 and 78.62 percent, respectively. We analyze the importance of phenotypic features and image features in different classification tasks. The importance of the structure MRI images is highlighted by incorporating phenotypic features with image features to generate hybrid features. We also visualize the features of three neuropsychiatric disorders and analyze their locations in the brain region.

Phylogenetic trees are unable to represent the evolutionary process for a collection of species if reticulation events happened, and a generalized model named phylogenetic network was introduced consequently. However, the representation of the evolutionary process for one gene is actually a phylogenetic tree that is ‘`contained’' in the phylogenetic network for the considered species containing the gene. Thus a fundamental computational problem named Tree Containment problem arises, which asks whether a phylogenetic tree is contained in a phylogenetic network. The previous research on the problem mainly focused on its rooted version of which the considered tree and network are rooted, and several algorithms were proposed when the considered network is binary or structure-restricted. There is almost no algorithm for its unrooted version except the recent fixed-parameter algorithm with runtime$O(4^kn^2)$, where k and n are the reticulation number and size of the considered unrooted binary phylogenetic network$N$, respectively. As the runtime is a little expensive when considering big values of k, we aim to improve it and successfully propose a fixed-parameter algorithm with runtime$O(2.594^kn^2)$in the paper. Additionally, we experimentally show its effectiveness on biological data and simulated data.

A drug-drug interaction (DDI) is defined as an association between two drugs where the pharmacological effects of a drug are influenced by another drug. Positive DDIs can usually improve the therapeutic effects of patients, but negative DDIs cause the major cause of adverse drug reactions and even result in the drug withdrawal from the market and the patient death. Therefore, identifying DDIs has become a key component of the drug development and disease treatment. In this study, we propose a novel method to predict DDIs based on the integrated similarity and semi-supervised learning (DDI-IS-SL). DDI-IS-SL integrates the drug chemical, biological and phenotype data to calculate the feature similarity of drugs with the cosine similarity method. The Gaussian Interaction Profile kernel similarity of drugs is also calculated based on known DDIs. A semi-supervised learning method (the Regularized Least Squares classifier) is used to calculate the interaction possibility scores of drug-drug pairs. In terms of the 5-fold cross validation, 10-fold cross validation and de novo drug validation, DDI-IS-SL can achieve the better prediction performance than other comparative methods. In addition, the average computation time of DDI-IS-SL is shorter than that of other comparative methods. Finally, case studies further demonstrate the performance of DDI-IS-SL in practical applications.

The identification of compound-protein relations (CPRs), which includes compound-protein interactions (CPIs) and compound-protein affinities (CPAs), is critical to drug development. A common method for compound-protein relation identification is the use of in vitro screening experiments. However, the number of compounds and proteins is massive, and in vitro screening experiments are labor-intensive, expensive, and time-consuming with high failure rates. Researchers have developed a computational field called virtual screening (VS) to aid experimental drug development. These methods utilize experimentally validated biological interaction information to generate datasets and use the physicochemical and structural properties of compounds and target proteins as input information to train computational prediction models. At present, deep learning has been widely used in computer vision and natural language processing and has experienced epoch-making progress. At the same time, deep learning has also been used in the field of biomedicine widely, and the prediction of CPRs based on deep learning has developed rapidly and has achieved good results. The purpose of this study is to investigate and discuss the latest applications of deep learning techniques in CPR prediction. First, we describe the datasets and feature engineering (i.e., compound and protein representations and descriptors) commonly used in CPR prediction methods. Then, we review and classify recent deep learning approaches in CPR prediction. Next, a comprehensive comparison is performed to demonstrate the prediction performance of representative methods on classical datasets. Finally, we discuss the current state of the field, including the existing challenges and our proposed future directions. We believe that this investigation will provide sufficient references and insight for researchers to understand and develop new deep learning methods to enhance CPR predictions.

Modern data centers host diverse applications, which generate a mix of short flows with stringent latency requirement and long flows requiring large sustained throughput. To solve the problem of resource competition between the mixed flows, we propose an adaptive traffic isolation scheme APS. Based on the packet spraying scheme in the multipath transmission, APS dynamically separates long flows from short ones on different paths to provide the low latency for the short flows. Meanwhile, to resolve the out-of-order problem, APS limits the long flows to a few paths with Equal Cost Multi Path (ECMP). Experimental results of NS2 simulation and testbed implementation show that, APS reduces the average completion time for short flows by up to 60 percent and increases the throughputs for long flows by about 1.68x over the state-of-the-art multipath transmission schemes.

Fractional repetition (FR) codes are a class of repair efficient erasure codes that can recover a failed storage node with both optimal repair bandwidth and complexity. In this paper, we study the minimum distance of FR codes, which is the smallest number of nodes whose failure leads to the unrecoverable loss of the stored file. We derive a new upper bound on the minimum distance of FR codes, which is tighter than the Singleton bound and a Singleton-like bound that takes locality into account. Based on regular graphs and combinatorial designs, several families of FR codes with optimal minimum distance are obtained.

Multi-view 3D reconstruction generally adopts the feature fusion strategy to guide the generation of 3D shape for objects with different views. Empirically, the correspondence learning of object regions across different views enables better feature fusion. However, such idea has not been fully exploited in existing methods. Furthermore, current methods fail to explore the intrinsic dependency among regions within a 3D shape, leading to a rough reconstruction result. To address the above issues, we propose a Dual-View 3D Point Cloud reconstruction architecture named DVPC, which takes two views images as inputs, and progressively generates a refined 3D point cloud. First, a point cloud generation network is assigned to generate a coarse point cloud for each input view. Second, a dual-view point clouds synthesis network is presented in DVPC. It constructs a regional attention mechanism to learn a high-quality correspondence among regions across two coarse point clouds in different views, so that our DVPC can achieve feature fusion accurately. And then it develops a point cloud deformation module to produce a relatively-precise point cloud via establishing the communication between the coarse point cloud and the fused feature. Lastly, a point-region transformer network is devised to model the dependency among regions within the relatively-precise point cloud. With the dependency, the relatively-precise point cloud is refined into a desirable 3D point cloud with rich details. Qualitative and quantitative experiments on the ShapeNet and Pix3D datasets demonstrate that the proposed DVPC outperforms the state-of-the-art methods in terms of reconstruction quality.

Effective fusion of multimodal magnetic resonance imaging (MRI) is of great significance to boost the accuracy of glioma grading thanks to the complementary information provided by different imaging modalities. However, how to extract the common and distinctive information from MRI to achieve complementarity is still an open problem in information fusion research. In this study, we propose a deep neural network model termed as multimodal disentangled variational autoencoder (MMD-VAE) for glioma grading based on radiomics features extracted from preoperative multimodal MRI images. Specifically, the radiomics features are quantized and extracted from the region of interest for each modality. Then, the latent representations of variational autoencoder for these features are disentangled into common and distinctive representations to obtain the shared and complementary data among modalities. Afterwards, cross-modality reconstruction loss and common-distinctive loss are designed to ensure the effectiveness of the disentangled representations. Finally, the disentangled common and distinctive representations are fused to predict the glioma grades, and SHapley Additive exPlanations (SHAP) is adopted to quantitatively interpret and analyze the contribution of the important features to grading. Experimental results on two benchmark datasets demonstrate that the proposed MMD-VAE model achieves encouraging predictive performance (AUC:0.9939) on a public dataset, and good generalization performance (AUC:0.9611) on a cross-institutional private dataset. These quantitative results and interpretations may help radiologists understand gliomas better and make better treatment decisions for improving clinical outcomes.

The stock market investors aim at maximizing their investment returns. Stock recommendation task is to recommend stocks with higher return ratios for the investors. Most stock prediction methods study the historical sequence patterns to predict stock trend or price in the near future. In fact, the future price of a stock is correlated not only with its historical price, but also with other stocks. In this article, we take into account the relationships between stocks (corporations) by stock relation graph. Furthermore, we propose a Time-aware Relational Attention Network (TRAN) for graph-based stock recommendation according to return ratio ranking. In TRAN, the time-aware relational attention mechanism is designed to capture time-varying correlation strengths between stocks by the interaction of historical sequences and stock description documents. With the dynamic strengths, the nodes of the stock relation graph aggregate the features of neighbor stock nodes by graph convolution operation. For a given group of stocks, the proposed TRAN model can output the ranking results of stocks according to their return ratios. The experimental results on several real-world datasets demonstrate the effectiveness of our TRAN for stock recommendation.

Matrix completion, whose goal is to recover a matrix from a few entries observed, is a fundamental model behind many applications. Our study shows that, in many applications, the to-be-complete matrix can be represented as the sum of a low-rank matrix and a sparse matrix associating with side information matrices. The low-rank matrix depicts the global patterns while the sparse matrix characterizes the local patterns, which are often described by the side information. Accordingly, to achieve high-quality matrix completion, we propose a Feature and Nuclear Norm Minimization (FNNM) model. The rationale of FNNM is to employ transductive completion to generalize the global pattern and inductive completion to recover the local pattern. Alternative minimization algorithm based on fixed-point iteration is developed to numerically solve the FNNM model. FNNM has demonstrated promising results on a variety of applications, including movie recommendation, drug-target interaction prediction, and multi-label learning, consistently outperforming the state-of-the-art matrix completion algorithms.

The accurate prediction of isocitrate dehydrogenase (IDH) mutation and glioma segmentation are important tasks for computer-aided diagnosis using preoperative multimodal magnetic resonance imaging (MRI). The two tasks are ongoing challenges due to the significant inter-tumor and intra-tumor heterogeneity. The existing methods to address them are mostly based on single-task approaches without considering the correlation between the two tasks. In addition, the acquisition of IDH genetic labels is expensive and costly, resulting in a limited number of IDH mutation data for modeling. To comprehensively address these problems, we propose a fully automated multimodal MRI-based multi-task learning framework for simultaneous glioma segmentation and IDH genotyping. Specifically, the task correlation and heterogeneity are tackled with a hybrid CNN-Transformer encoder that consists of a convolutional neural network and a transformer to extract the shared spatial and global information learned from a decoder for glioma segmentation and a multi-scale classifier for IDH genotyping. Then, a multi-task learning loss is designed to balance the two tasks by combining the segmentation and classification loss functions with uncertain weights. Finally, an uncertainty-aware pseudo-label selection is proposed to generate IDH pseudo-labels from larger unlabeled data for improving the accuracy of IDH genotyping by using semi-supervised learning. We evaluate our method on a multi-institutional public dataset. Experimental results show that our proposed multi-task network achieves promising performance and outperforms the single-task learning counterparts and other existing state-of-the-art methods. With the introduction of unlabeled data, the semi-supervised multi-task learning framework further improves the performance of glioma segmentation and IDH genotyping. The source codes of our framework are publicly available at https://github.com/miacsu/MTTU-Net.git.

The video streaming system employs adaptive bitrate (ABR) algorithms to optimize a user’s quality of experience. However, it is hard for ABR algorithms to choose the right bitrate consistently under highly dynamic bandwidth fluctuations in wild Internet. In this article, we propose a building block on the client side named Opportunistic Chunk Replacement Mechanism (OCRM) to help existing ABR algorithms make full use of the available bandwidth to improve the network utilization and viewing experience of users. Specifically, the servers take advantages of the spare bandwidth to opportunistically transmit high-quality chunks (called opportunistic chunks ) with low priority to the client, without incurring any extra delay. Then, the client player replaces the low-quality chunks with the opportunistic ones that have high quality. We compare OCRM with state-of-the-art ABR algorithms by using trace-driven experiments spanning a wide variety of quality of experience metrics and network conditions. The test results show that OCRM effectively achieves high network utilization and improves the user’s viewing experience by up to 35%.

Modern data centers often host multiple applications with diverse network demands. To provide fair bandwidth allocation to several thousand traversing flows, Approximate Fair Queueing (AFQ) utilizes multiple priority queues in switch to approximate ideal fair queueing. However, due to limited number of queues in programmable switches, AFQ easily experiences high packet loss and low link utilization. In this paper, we propose Elastic Fair Queueing (EFQ), which leverages limited priority queues to flexibly achieve both high network utilization and fair bandwidth allocation. EFQ dynamically assigns the free buffer space in priority queues for each packet to obtain high utilization without sacrificing flow-level fairness. The results of simulation experiments and real implementations show that EFQ reduces the average flow completion time by up to 82% over the state-of-the-art fair bandwidth allocation mechanisms.

With the popularity of social media platforms, patients tend to share their experiences and opinions on them, and patient feedback is key to improving health services. Sentiment analysis techniques have been applied to automatically analyze the patients’ opinions to understand the quality of healthcare. Aspect extraction, which aims to identify the opinion targets in the text, is an important step towards understanding the patient’s opinion towards particular target or entity. However, due to the complex nature of medical domain data, existing approaches take much execution time. To address this, we presents a new approach for aspect extraction and refinement to smooth the sentiment analysis process. Furthermore, this work also introduces an intelligent weighting scheme for classifying the final aspects. For the experimental evaluation, a dataset from Yelp and RateMDs has been utilized. Experimental results show that the proposed model outperforms existing methods.

Predicting the future diagnoses from patients’ historical Electronic Health Records (EHR) is a significant task in healthcare. EHR consist of multiple modal data, each modality has different features and contains a wealth of information of patients. However, most of the existing EHR-based prediction methods either only use unimodal data, or fail to fully explore the correlation between different modalities when fusing multimodal data. To address these challenges, we propose a Multimodal Attention-based fusIon Networks (MAIN) for diagnosis prediction. In this model, we first design different feature extraction modules for each modality. Then, an inter-modal correlation module which contains two layers is applied to capture the intermodal correlation. Finally, a multimodal fusion module based on weighted averaging is utilized to integrate the representations derived from different modalities and their correlation to obtain the patient representation for diagnosis prediction. We evaluate our proposed model on two medical datasets, and the experimental results demonstrate the effectiveness of MAIN.

Medical image segmentation is a fundamental step for diagnosis and prognosis. This study proposes a new body and edge aware network for automated 2D medical image segmentation (called BEA-SegNet). The proposed BEA-SegNet consists of a shared encoder, a body and edge decouple (BEdecouple) module, two parallel decoders for body and edge segmentation. In the encoder and decoders, short-term multi-scale concatenation (STMSC) modules are utilized to implement multi-scale representation. We design a BEdecouple module to decouple the convolutional features into the body and edge features, making the proposed method be body and edge aware. The body and edge decoders utilize Bedecouple modules in each level to learn more effective features for the body and edge segmentation respectively, and their outputs are fused to generate the final segmentation. Besides, the body and edge supervision are applied to improve the final segmentation. The proposed BEA-SegNet is trained and evaluated on the International Skin Imaging Collaboration challenge 2018 dataset (ISIC2018). Experimental results show that the proposed BEA-SegNet achieves an average Dice similarity coefficient of 90.3% and an average Hausdorff distance of 15.9 for the skin lesion segmentation task and outperforms five benchmarks for skin lesion segmentation.

Automatic detection of Alzheimer’s disease (AD) is conducive to intervention in the disease progression. MMSE score prediction can reveal the development of AD. In recent years, some studies have designed multi-task methods for AD detection and MMSE score prediction to take advantage of the correlation between them. However, how to use the correlation between the two task features is still a problem. To address this challenge, we propose a multi-task feature interactive leanrning network (MTFIL-Net) to perform AD detection and MMSE score prediction. First, we interact the features acquired by CNNs corresponding to the two tasks to take advantage of the feature correlation between the two tasks. The interaction module extracts the shared features of the two tasks and concatenate them with the features of the two task. Then, we design a joint loss based on cross entropy and smooth L1 function. We use the distribution of MMSE scores to dynamically adjust the relationship between the two tasks. We validate our method with subjects from the Alzheimer’s Disease Neuroimaging Initiative (ADNI). We use the ADNI1 dataset for training and testing, and used the ADNI2 dataset as an external validation set. Our proposed MTFIL-Net reached an ACC of 0.86 for AD detection and a correlation coefficient of 0.67 for MMSE score prediction on the ADNI1 dataset, and reached an ACC of 0.85 for AD detection and a correlation coefficient of 0.66 for MMSE score prediction on the ADNI2 dataset. Experiment results show that MTFIL-Net effectively utilizes the correlation between AD and MMSE score.

Single-cell RNA sequencing (scRNA-seq) presents cell heterogeneity in a high resolution to explore cell development. The high dimension and the high noise in scRNAseq data bring some computational challenges. By introducing optimal transmission regularization, we proposed a novel deep clustering method, called SCOTCluster, which accurately learned low-dimensional representations. In SCOTCluster, a joint training strategy was designed by integrating AutoEncoder and soft k-means. Notably, to improve simultaneously the accuracy and robustness, the optimal transmission was introduced in the objective function of soft k-means, and entropy regularization and Sinkhorn iterative algorithm were performed to constraint the cluster size. To test the performance, we compared SCOTCluster with five state-of the-art methods on 16 real large-scale scRNA-seq datasets. The experimental results showed that SCOTCluster improved the training stability and clustering performance.

Automatic identification of adventitious respiratory sound has still been a challenging problem in recent years. To address this challenge, we propose an adventitious respiratory sound classification network (ARSC-Net), which combines residual block with channel-spatial attention for accurate classification. Specifically, we extract two types of features from adventitious respiratory sound, including Mel-Frequency Cepstral Coefficients (MFCCs) and Mel-spectrogram. The two types of features are entered into the parallel encoders paths with residual attention for extracting feature representation, and then fused into a channel-spatial attention module to adaptively focus on the important features between channel and spatial part for the classification task. Moreover, the channel-spatial attention can enhance the feature representation, in which the channel attention explores the inter-channel relationship of the spectrums, and then the inter-spatial correlation mapping is generated by the spatial attention introduced serially. We evaluate our proposed method on ICBHI 2017 database. Experimental results show that our proposed method achieves encouraging predictive performance with an accuracy of 80.0% for identifying abnormal sounds from normal sounds, and with an accuracy of 92.4% for distinguishing crackles from wheezes. In addition, our method also achieves a score of 56.76% for the four-class sound classification of adventitious sounds and outperforms several state-of-the-art methods.

Alternative splicing has a significant impact in the process of gene regulation and it greatly raises the complexity of protein species and functions. At present, the gene function has been extensively studied, and there are multiple public databases of gene function annotations. Due to the high cost and the large number of isoforms, it is currently impossible to conduct extensive experimental analysis on isoform functions. Functions at the level of splicing isoforms remain largely unknown. Computational prediction provides an alternative to functional annotation of splice isoforms. Here we present GraphIsoFun, a novel computational approach to predict isoform function by integrating both gene and isoform-level expression information with graph neural network. It first constructs a heterogeneous co-expression network with both genes and isoforms being nodes. This network integrates the rich association between isoforms and genes, and provides a bridge for connecting isoform and gene functions. Then GraphIsoFun uses the graph neural network to mine the information of the co-expression network and obtains the final prediction score of isoforms for each function. The experiments on multiple datasets show that GraphIsoFun surpasses existing methods based on different evaluation indicators.

Viral infectious diseases have become a serious threat to human health and global security. The receptor-binding is the first step of viral infection. Identifying potential virus-receptor interactions gives us a new perspective on understanding interaction mechanisms, and further find potential targets for prevention and therapy of viral infectious diseases. Many predicted methods have been proposed to identify potential virus-receptor interactions. They didn’t focus on fusing multiple biological features and using multiple similarity measures depending on multiple biological characteristics to improve the prediction performance. In this study, a novel predictive method (LOMCVRI) is proposed to identify potential VirusReceptor Interactions based on similarity fusion and Matrix Completion of Linear optimization. In LOMCVRI, the viral protein structure similarity, the viral protein sequence similarity and the viral genomic sequence similarity are computed based on viral secondary structure features, viral protein sequences and viral genomic sequences, respectively. The gaussian radial basis function is used to improve these viral similarities. They are further combined into a compositive viral similarity by the linear weighting average method. Second, we compute the receptor conserved domain sequence similarity, the receptor sequence similarity and the receptor protein-protein interaction network (PPI) similarity based on the receptor conserved domain sequences, the receptor amino acid sequences and the human PPI data. They are also fused into a compositive receptor similarity. Next, the K-nearest neighbor preprocessing is employed to prefill missing entries in original interaction matrix. Finally, based on updated virus-receptor interactions and compositive similarities of virus and receptor, the matrix completion of linear optimization is applied to identify potential virus-receptor interactions. 10-fold Cross-Validation (10CV) and Leave-One-Out Cross-Validation (LOOCV) experimental results show that the area under of ROC curve (AUC) values of LOMCVRI are 0.9231 and 0.9438, respectively, which consistently outperforms all other related methods. In addition, a case study further confirms the effectiveness of our method.

Epidermal growth factor receptor (EGFR) genotyping is essential to treatment guidelines for the use of tyrosine kinase inhibitors in lung adenocarcinoma. However, accurate and noninvasive methods to detect the EGFR gene are ongoing challenges. In this study, we propose a hybrid framework, namely HC-DLR, to noninvasively predict EGFR mutation status by fusing multi-source features including low-level handcrafted radiomics (HCR) features, high-level deep learning-based radiomics (DLR) features, and demographics features. The HCR features first are selected from massive handcrafted features extracted from CT images. The DLR features are also extracted from CT images using the pre-trained 3D DenseNet. Then, multi-source feature representations are refined and fused to build an HC-DLR model for improving the predictive performance of EGFR mutations. The proposed method is evaluated on a newly collected dataset with 670 patients. Experimental results show that the HC-DLR model achieves an encouraging predictive performance with an AUC of 0.76, an accuracy of 72.47%, and an F1-score of 71.35%, which may have potential clinical value for predicting EGFR mutations in lung adenocarcinoma.

Nowadays, the distributed key-value stores have become the basic building block for large scale cloud applications. In large-scale distributed key-value stores, many key-value access operations, which will be processed in parallel on different servers, are usually generated for the data required by a single end-user request. Hence, the completion time of the end request is determined by the last completed key-value access operation. Accordingly, scheduling the order of key-value access operations of different end requests can effectively reduce their completion time, improving the user experience. However, existing algorithms are either hard to employ in distributed key-value stores due to the relatively large cooperation overhead for centralized information or unable to adapt to the time-varying load and server performance under different traffic patterns. In this paper, we first formalize the scheduling problem for small mean request completion time. As a step further, because of the NP-hardness of this problem, we heuristically design the distributed adaptive scheduler (DAS) for distributed key-value stores. DAS reduces the average request completion time by a distributed combination of the largest remaining processing time last and shortest remaining process time first algorithms. Moreover, DAS is adaptive to the time-varying server load and performance. Extensive simulations show that DAS reduces the mean request completion time by more than 15 ~ 50% compared to the default first come first served algorithm and outperforms the existing Rein-SBF algorithm under various scenarios.

Explicit Congestion Notification (ECN) is widely utilized in modern data centers to achieve low latency and high throughput for various applications. In recent years, however, even with the sustainable growth of link bandwidth in data centers, the switch buffer size does not increase remarkably. Consequently, the standard per-port ECN scheme suffers from excessive packet loss. Though the shared-buffer ECN scheme alleviates the packet loss, we observe that it leads to severe unfairness, which we term as the Port Starvation problem. When flows destined for some ports have aggressively occupied the shared buffer, the later-arrival flows destined for other ports will be ECN-marked unfairly and obtain significantly lower throughput. To address the port starvation problem, we design a buffer-aware fair ECN-marking (BFEM) scheme for shallow-buffered switch. BFEM leverages the shared buffer to reduce packet loss and meanwhile punishes aggressive flows by ECN marking. We evaluate BFEM with both 40Gbps P4 testbed implementation and large-scale NS2 simulation. The test results show that, by improving fairness between egress ports, BFEM increases total link utilization and reduces the average flow completion time by up to 40% compared with the state-of-the-art per-port and shared-buffer ECN marking schemes.

To improve the quality of experience for worldwide users, an increasing number of service providers deploy their services on geographically dispersed data centers, which are connected by wide area network (WAN). In the cross-datacenter networks, however, the intra- and inter-datacenter parts have different characteristics, including switch buffer depth, round-trip time and bandwidth. Besides, most of intra-DC flows belong to interactive services that require low delay while inter-DC flows typically need to achieve high throughput. Unfortunately, existing sender-based and receiver-driven transport protocols do not consider the network heterogeneity between inter- and intra- DC networks so that they fail to simultaneously achieve low latency for intra-DC flows and high throughput for inter-DC flows. This paper proposes a general hybrid congestion control mechanism called GTCP to address this problem. When the inter-DC flow detects congestion inside data center, it switches to the receiver-driven mode to avoid the impact on intra-DC flows. Otherwise, it switches back to the sender-based mode to proactively explore the available bandwidth. Besides, the intra-DC flow leverages the pausing mechanism to eliminate the queue build-up. Through a series of testbed experiments and large-scale NS2 simulations, we demonstrate that GTCP reduces flow completion time by up to 79.3% compared with existing protocols.

Modern datacenter applications bring fundamental challenges to transport protocols as they simultaneously require low latency and high throughput. Recent receiver-driven trans-port protocols transmit only one data packet once receiving each grant or credit packet from the receiver to achieve ultra-low queueing delay and zero packet loss. However, the round-trip time variation and the highly dynamic background traffic significantly deteriorate the performance of receiver-driven transport protocols, resulting in under-utilized bandwidth. This paper designs a simple yet effective solution called RPO that retains the advantages of receiver-driven transmission while efficiently utilizing the available bandwidth. Specifically, RPO rationally uses low-priority opportunistic packets to ensure high network utilization without increasing the queueing delay of high-priority normal packets. In addition, since RPO only uses Explicit Congestion Notification (ECN) marking function and priority queues, RPO is ready to deploy on switches. We implement RPO in Linux hosts with DPDK. Our small-scale testbed experiments and large-scale simulations show that RPO significantly improves the network utilization by up to 35% under high workload over the state-of-the-art receiver-driven transmission schemes, without introducing additional queueing delay.

In this paper, we study the following Minimum Cost Multi-Covering (MCMC) problem: Given a set of n client points C and a set of m server points S in a fixed dimensional ℝd space, determine a set of disks centered at these server points so that each client point c is covered by at least k(c) disks and the total cost of these disks is minimized, where k(-) is a function that maps every client point to some non-negative integer no more than m and the cost of each disk is measured by the α-th power of its radius for some constant α > 0. MCMC is a fundamental optimization problem with applications in many areas such as wireless/sensor networking. Despite extensive research on this problem in the past two decades, only constant approximations were known for general k. It has been an open problem for a long time to determine whether a PTAS is possible. In this paper, we give an affirmative answer to this question by presenting the first PTAS for it. Our approach is based on a number of novel techniques, such as Balanced Recursive Realization and Bubble Charging, and new insights to the problem which are somewhat counter-intuitive. Particularly, we show that instead of optimizing each disk as a whole, it is possible to further approximate each disk with a set of sub-boxes and optimize them at the sub-disk level. This allows us to first compute an approximate disk cover with minimum cost through dynamic programming, and then obtain the desired disk cover through a balanced recursive realization procedure. Our techniques have the potential to be used to other geometric (covering) problems.

The Novel Coronavirus Disease 2019 (COVID-19) has become an international public health emergency, which poses the most serious threat to the human health around the world. Accumulating evidences have shown that the new coronavirus could not only infect human beings, but also can infect other species which might result in the cross-species infections. In this research, 1056 ACE2 protein sequences are collected from the NCBI database, and 173 species with >60% sequence identity compared with that of human beings are selected for further analysis. We find 14 polar residues forming the binding interface of ACE2/2019-nCoV-Spike complex play an important role in maintaining protein-protein stability. Among them, 8 polar residues at the same positions with that of human ACE2 are highly conserved, which ensure its basic binding affinity with the novel coronavirus. 5 of other 6 unconserved polar residues (positions at human ACE2: Q24, D30, K31, H34 and E35) are proved to have an effect on the binding patterns among species. We select 21 species keeping close contacts with human beings, construct their ACE2 three-dimensional structures by Homology Modeling method and calculate the binding free energies of their ACE2/2019-nCoV-Spike complexes. We find the ACE2 from all the 21 species possess the capabilities to bind with the novel coronavirus. Compared with the human beings, 8 species (cow, deer, cynomys, chimpanzee, monkey, sheep, dolphin and whale) present almost the same binding abilities, and 3 species (bat, pig and dog) show significant improvements in binding affinities. We hope this research could provide significant help for the future epidemic detection, drug and vaccine development and even the global eco-system protections.

Drug repositioning can drastically decrease the cost and duration taken by traditional drug research and development while avoiding the occurrence of unforeseen adverse events. With the rapid advancement of high-throughput technologies and the explosion of various biological data and medical data, computational drug repositioning methods have been appealing and powerful techniques to systematically identify potential drug-target interactions and drug-disease interactions. In this review, we first summarize the available biomedical data and public databases related to drugs, diseases and targets. Then, we discuss existing drug repositioning approaches and group them based on their underlying computational models consisting of classical machine learning, network propagation, matrix factorization and completion, and deep learning based models. We also comprehensively analyze common standard data sets and evaluation metrics used in drug repositioning, and give a brief comparison of various prediction methods on the gold standard data sets. Finally, we conclude our review with a brief discussion on challenges in computational drug repositioning, which includes the problem of reducing the noise and incompleteness of biomedical data, the ensemble of various computation drug repositioning methods, the importance of designing reliable negative samples selection methods, new techniques dealing with the data sparseness problem, the construction of large-scale and comprehensive benchmark data sets and the analysis and explanation of the underlying mechanisms of predicted interactions.

Advances in sequencing technologies facilitate personalized disease-risk profiling and clinical diagnosis. In recent years, some great progress has been made in noninvasive diagnoses based on cell-free DNAs (cfDNAs). It exploits the fact that dead cells release DNA fragments into the circulation, and some DNA fragments carry information that indicates their tissues-of-origin (TOOs). Based on the signals used for identifying the TOOs of cfDNAs, the existing methods can be classified into three categories: cfDNA mutation-based methods, methylation pattern-based methods and cfDNA fragmentation pattern-based methods. In cfDNA mutation-based methods, the SNP information or the detected mutations in driven genes of certain diseases are employed to identify the TOOs of cfDNAs. Methylation pattern-based methods are developed to identify the TOOs of cfDNAs based on the tissue-specific methylation patterns. In cfDNA fragmentation pattern-based methods, cfDNA fragmentation patterns, such as nucleosome positioning or preferred end coordinates of cfDNAs, are used to predict the TOOs of cfDNAs. In this paper, the strategies and challenges in each category are reviewed. Furthermore, the representative applications based on the TOOs of cfDNAs, including noninvasive prenatal testing, noninvasive cancer screening, transplantation rejection monitoring and parasitic infection detection, are also reviewed. Moreover, the challenges and future work in identifying the TOOs of cfDNAs are discussed. Our research provides a comprehensive picture of the development and challenges in identifying the TOOs of cfDNAs, which may benefit bioinformatics researchers to develop new methods to improve the identification of the TOOs of cfDNAs.

Various microbes have proved to be closely related to the pathogenesis of human diseases. While many computational methods for predicting human microbe-disease associations (MDAs) have been developed, few systematic reviews on these methods have been reported. In this study, we provide a comprehensive overview of the existing methods. Firstly, we introduce the data used in existing MDA prediction methods. Secondly, we classify those methods into different categories by their nature and describe their algorithms and strategies in detail. Next, experimental evaluations are conducted on representative methods using different similarity data and calculation methods to compare their prediction performances. Based on the principles of computational methods and experimental results, we discuss the advantages and disadvantages of those methods and propose suggestions for the improvement of prediction performances. Considering the problems of the MDA prediction at present stage, we discuss future work from three perspectives including data, methods and formulations at the end.

With the development of high-throughput technology and the accumulation of biomedical data, the prior information of biological entity can be calculated from different aspects. Specifically, drug-drug similarities can be measured from target profiles, drug-drug interaction and side effects. Similarly, different methods and data sources to calculate disease ontology can result in multiple measures of pairwise disease similarities. Therefore, in computational drug repositioning, developing a dynamic method to optimize the fusion process of multiple similarities is a crucial and challenging task. In this study, we propose a multi-similarities bilinear matrix factorization (MSBMF) method to predict promising drug-associated indications for existing and novel drugs. Instead of fusing multiple similarities into a single similarity matrix, we concatenate these similarity matrices of drug and disease, respectively. Applying matrix factorization methods, we decompose the drug-disease association matrix into a drug-feature matrix and a disease-feature matrix. At the same time, using these feature matrices as basis, we extract effective latent features representing the drug and disease similarity matrices to infer missing drug-disease associations. Moreover, these two factored matrices are constrained by non-negative factorization to ensure that the completed drug-disease association matrix is biologically interpretable. In addition, we numerically solve the MSBMF model by an efficient alternating direction method of multipliers algorithm. The computational experiment results show that MSBMF obtains higher prediction accuracy than the state-of-the-art drug repositioning methods in cross-validation experiments. Case studies also demonstrate the effectiveness of our proposed method in practical applications. Availability: The data and code of MSBMF are freely available at https://github.com/BioinformaticsCSU/MSBMF. Corresponding author: Jianxin Wang, School of Computer Science and Engineering, Central South University, Changsha, Hunan 410083, P. R. China. E-mail: [email protected] Supplementary Data: Supplementary data are available online at https://academic.oup.com/bib.

Identifying the frequencies of the drug-side effects is a very important issue in pharmacological studies and drug risk-benefit. However, designing clinical trials to determine the frequencies is usually time consuming and expensive, and most existing methods can only predict the drug-side effect existence or associations, not their frequencies. Inspired by the recent progress of graph neural networks in the recommended system, we develop a novel prediction model for drug-side effect frequencies, using a graph attention network to integrate three different types of features, including the similarity information, known drug-side effect frequency information and word embeddings. In comparison, the few available studies focusing on frequency prediction use only the known drug-side effect frequency scores. One novel approach used in this work first decomposes the feature types in drug-side effect graph to extract different view representation vectors based on three different type features, and then recombines these latent view vectors automatically to obtain unified embeddings for prediction. The proposed method demonstrates high effectiveness in 10-fold cross-validation. The computational results show that the proposed method achieves the best performance in the benchmark dataset, outperforming the state-of-the-art matrix decomposition model. In addition, some ablation experiments and visual analyses are also supplied to illustrate the usefulness of our method for the prediction of the drug-side effect frequencies. The codes of MGPred are available at https://github.com/zhc940702/MGPred and https://zenodo.org/record/4449613.

Proteins are dominant executors of living processes. Compared to genetic variations, changes in the molecular structure and state of a protein (i.e. proteoforms) are more directly related to pathological changes in diseases. Characterizing proteoforms involves identifying and locating primary structure alterations (PSAs) in proteoforms, which is of practical importance for the advancement of the medical profession. With the development of mass spectrometry (MS) technology, the characterization of proteoforms based on top-down MS technology has become possible. This type of method is relatively new and faces many challenges. Since the proteoform identification is the most important process in characterizing proteoforms, we comprehensively review the existing proteoform identification methods in this study. Before identifying proteoforms, the spectra need to be preprocessed, and protein sequence databases can be filtered to speed up the identification. Therefore, we also summarize some popular deconvolution algorithms, various filtering algorithms for improving the proteoform identification performance and various scoring methods for localizing proteoforms. Moreover, commonly used methods were evaluated and compared in this review. We believe our review could help researchers better understand the current state of the development in this field and design new efficient algorithms for the proteoform characterization.

The rapid increase of genome data brought by gene sequencing technologies poses a massive challenge to data processing. To solve the problems caused by enormous data and complex computing requirements, researchers have proposed many methods and tools which can be divided into three types: big data storage, efficient algorithm design and parallel computing. The purpose of this review is to investigate popular parallel programming technologies for genome sequence processing. Three common parallel computing models are introduced according to their hardware architectures, and each of which is classified into two or three types and is further analyzed with their features. Then, the parallel computing for genome sequence processing is discussed with four common applications: genome sequence alignment, single nucleotide polymorphism calling, genome sequence preprocessing, and pattern detection and searching. For each kind of application, its background is firstly introduced, and then a list of tools or algorithms are summarized in the aspects of principle, hardware platform and computing efficiency. The programming model of each hardware and application provides a reference for researchers to choose high-performance computing tools. Finally, we discuss the limitations and future trends of parallel computing technologies.

MOTIVATION: Compared with the second-generation sequencing technologies, the third-generation sequencing technologies allows us to obtain longer reads (average ∼10 kbps, maximum 900 kbps), but brings a higher error rate (∼15% error rate). Nanopolish is a variant and methylation detection tool based on hidden Markov model, which uses Oxford Nanopore sequencing data for signal-level analysis. Nanopolish can greatly improve the accuracy of assembly, whereas it is limited by long running time since most executive parts of Nanopolish is a serial and computationally expensive process. RESULTS: In this paper, we present an effective polishing tool, Multithreading Nanopolish (MultiNanopolish), which decomposes the whole process of iterative calculation in Nanopolish into small independent calculation tasks, making it possible to run this process in the parallel mode. Experimental results show that MultiNanopolish reduces running time by 50% with read-uncorrected assembler (Miniasm) and 20% with read-corrected assembler (Canu and Flye) based on 40 threads mode compared to the original Nanopolish. AVAILABILITY AND IMPLEMENTATION: MultiNanopolish is available at GitHub: https://github.com/BioinformaticsCSU/MultiNanopolish. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: Oxford Nanopore sequencing producing long reads at low cost has made many breakthroughs in genomics studies. However, the large number of errors in Nanopore genome assembly affect the accuracy of genome analysis. Polishing is a procedure to correct the errors in genome assembly and can improve the reliability of the downstream analysis. However, the performances of the existing polishing methods are still not satisfactory. RESULTS: We developed a novel polishing method, NeuralPolish, to correct the errors in assemblies based on alignment matrix construction and orthogonal Bi-GRU networks. In this method, we designed an alignment feature matrix for representing read-to-assembly alignment. Each row of the matrix represents a read, and each column represents the aligned bases at each position of the contig. In the network architecture, a bi-directional GRU network is used to extract the sequence information inside each read by processing the alignment matrix row by row. After that, the feature matrix is processed by another bi-directional GRU network column by column to calculate the probability distribution. Finally, a CTC decoder generates a polished sequence with a greedy algorithm. We used five real datasets and three assembly tools including Wtdbg2, Flye and Canu for testing, and compared the results of different polishing methods including NeuralPolish, Racon, MarginPolish, HELEN and Medaka. Comprehensive experiments demonstrate that NeuralPolish achieves more accurate assembly with fewer errors than other polishing methods and can improve the accuracy of assembly obtained by different assemblers. AVAILABILITY AND IMPLEMENTATION: https://github.com/huangnengCSU/NeuralPolish.git. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: High resolution annotation of gene functions is a central goal in functional genomics. A single gene may produce multiple isoforms with different functions through alternative splicing. Conventional approaches, however, consider a gene as a single entity without differentiating these functionally different isoforms. Towards understanding gene functions at higher resolution, recent efforts have focused on predicting the functions of isoforms. However, the performance of existing methods is far from satisfactory mainly because of the lack of isoform-level functional annotation. RESULTS: We present IsoResolve, a novel approach for isoform function prediction, which leverages the information from gene function prediction models with domain adaptation (DA). IsoResolve treats gene-level and isoform-level features as source and target domains, respectively. It uses DA to project the two domains into a latent variable space in such a way that the latent variables from the two domains have similar distribution, which enables the gene domain information to be leveraged for isoform function prediction. We systematically evaluated the performance of IsoResolve in predicting functions. Compared with five state-of-the-art methods, IsoResolve achieved significantly better performance. IsoResolve was further validated by case studies of genes with isoform-level functional annotation. AVAILABILITY AND IMPLEMENTATION: IsoResolve is freely available at https://github.com/genemine/IsoResolve. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: Emerging studies indicate that circular RNAs (circRNAs) are widely involved in the progression of human diseases. Due to its special structure which is stable, circRNAs are promising diagnostic and prognostic biomarkers for diseases. However, the experimental verification of circRNA-disease associations is expensive and limited to small-scale. Effective computational methods for predicting potential circRNA-disease associations are regarded as a matter of urgency. Although several models have been proposed, over-reliance on known associations and the absence of characteristics of biological functions make precise predictions are still challenging. RESULTS: In this study, we propose a method for predicting CircRNA-disease associations based on sequence and ontology representations, named CDASOR, with convolutional and recurrent neural networks. For sequences of circRNAs, we encode them with continuous k-mers, get low-dimensional vectors of k-mers, extract their local feature vectors with 1D CNN and learn their long-term dependencies with bi-directional long short-term memory. For diseases, we serialize disease ontology into sentences containing the hierarchy of ontology, obtain low-dimensional vectors for disease ontology terms and get terms' dependencies. Furthermore, we get association patterns of circRNAs and diseases from known circRNA-disease associations with neural networks. After the above steps, we get circRNAs' and diseases' high-level representations, which are informative to improve the prediction. The experimental results show that CDASOR provides an accurate prediction. Importing the characteristics of biological functions, CDASOR achieves impressive predictions in the de novo test. In addition, 6 of the top-10 predicted results are verified by the published literature in the case studies. AVAILABILITY AND IMPLEMENTATION: The code and data of CDASOR are freely available at https://github.com/BioinformaticsCSU/CDASOR.

MOTIVATION: Emerging evidence presents that traditional drug discovery experiment is time-consuming and high costs. Computational drug repositioning plays a critical role in saving time and resources for drug research and discovery. Therefore, developing more accurate and efficient approaches is imperative. Heterogeneous graph inference is a classical method in computational drug repositioning, which not only has high convergence precision, but also has fast convergence speed. However, the method has not fully considered the sparsity of heterogeneous association network. In addition, rough similarity measure can reduce the performance in identifying drug-associated indications. RESULTS: In this article, we propose a heterogeneous graph inference with matrix completion (HGIMC) method to predict potential indications for approved and novel drugs. First, we use a bounded matrix completion (BMC) model to prefill a part of the missing entries in original drug-disease association matrix. This step can add more positive and formative drug-disease edges between drug network and disease network. Second, Gaussian radial basis function (GRB) is employed to improve the drug and disease similarities since the performance of heterogeneous graph inference more relies on similarity measures. Next, based on the updated drug-disease associations and new similarity measures of drug and disease, we construct a novel heterogeneous drug-disease network. Finally, HGIMC utilizes the heterogeneous network to infer the scores of unknown association pairs, and then recommend the promising indications for drugs. To evaluate the performance of our method, HGIMC is compared with five state-of-the-art approaches of drug repositioning in the 10-fold cross-validation and de novo tests. As the numerical results shown, HGIMC not only achieves a better prediction performance but also has an excellent computation efficiency. In addition, cases studies also confirm the effectiveness of our method in practical application. AVAILABILITYAND IMPLEMENTATION: The HGIMC software and data are freely available at https://github.com/BioinformaticsCSU/HGIMC, https://hub.docker.com/repository/docker/yangmy84/hgimc and http://doi.org/10.5281/zenodo.4285640. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: The Anatomical Therapeutic Chemical (ATC) system is an official classification system established by the World Health Organization for medicines. Correctly assigning ATC classes to given compounds is an important research problem in drug discovery, which can not only discover the possible active ingredients of the compounds, but also infer theirs therapeutic, pharmacological and chemical properties. RESULTS: In this article, we develop an end-to-end multi-label classifier called CGATCPred to predict 14 main ATC classes for given compounds. In order to extract rich features of each compound, we use the deep Convolutional Neural Network and shortcut connections to represent and learn the seven association scores between the given compound and others. Moreover, we construct the correlation graph of ATC classes and then apply graph convolutional network on the graph for label embedding abstraction. We use all label embedding to guide the learning process of compound representation. As a result, by using the Jackknife test, CGATCPred obtain reliable Aiming of 81.94%, Coverage of 82.88%, Accuracy 80.81%, Absolute True 76.58% and Absolute False 2.75%, yielding significantly improvements compared to exiting multi-label classifiers. AVAILABILITY AND IMPLEMENTATION: The codes of CGATCPred are available at https://github.com/zhc940702/CGATCPred and https://zenodo.org/record/4552917.

To achieve security in wireless sensor networks (WSNs), it is important to be able to encrypt messages sent among sensor nodes by using shared keys between them. Due to resource constraints, achieving such key agreement in WSNs is nontrivial. Previous research indicates that key management schemes using deployment knowledge can significantly improve the performance of WSNs. Nevertheless, in these schemes, resilient local connectivity and resilient global connectivity become unstable when deployment error changes. To resolve the above problem, in this article, a one-way associated key management model is proposed. In this model, the key pool consists of two layers: 1) the global layer and 2) the local layer. According to different deployment errors, the number of keys allocated from the global key pool and local key pools can be dynamically adjusted, thereby improving the stability of networks' performance. In multiphase sensor networks, analysis and simulation indicate that our scheme has better adaptability in applications where deployment error changes as compared with related schemes.

For massive access to the Internet of Things, edge computing servers are installed on cellular ground base stations (GBSs) with fixed geographical locations, which easily suffer from traffic overload of the end user (EU) with high density and mobility. To provide reliable and flexible offloading service, unmanned aerial vehicles (UAV) are explored to assist edge computing, which relieves the computation offloading pressure of both EUs and GBS. However, most existing UAV researches focus on trajectory design to reduce offloading delay, which ignoring the variability of user distribution and the energy limitation of UAV. This article proposes a novel UAV-assisted edge computing framework, named as HOTSPOT, which locates the UAV in 3-D space according to the time-varying hot spot of user distribution and provides the corresponding edge computing offloading assistance. By formulating the UAV positioning problem into a maximum clique problem, a light-weighted deterministic algorithm is proposed based on stochastic gradient descent to search the optimal location of UAV. With the elaborate UAV position, HOTSPOT further gives an opportunistic offloading balanced scheme to reach low latency. Simulation results show that when the GBS load is 75%, HOTSPOT reduces the average offloading delay by 33%. When the GBS load reaches 90%, the average delay reduction is up to 80%.

Mobile-edge computing (MEC) is expected to provide reliable and low-latency computation offloading for massive Internet of Things (IoT) with the next generation networks, such as the sixth-generation (6G) network. However, the successful implementation of 6G depends on network densification, which brings new offloading challenges for edge computing, one of which is how to make offloading decisions facing densified servers considering both channel interference and queuing, which is an NP-hard problem. This article proposes a distributed-two-stage offloading (DTSO) strategy to give tradeoff solutions. In the first stage, by introducing the queuing theory and considering channel interference, a combinatorial optimization problem is formulated to calculate the offloading probability of each station. In the second stage, the original problem is converted to a nonlinear optimization problem, which is solved by a designed sequential quadratic programming (SQP) algorithm. To make an adjustable tradeoff between the latency and energy requirement among heterogeneous applications, an elasticity parameter is specially designed in DTSO. Simulation results show that compared to the latest works, DTSO can effectively reduce latency and energy consumption and achieve a balance between them based on application preferences.

Vital signs (e.g., respiration rate or heartbeat rate) sensing is of great importance to implement pervasive in-home healthcare. Traditional vital signs monitoring approaches usually require users to wear some dedicated sensors. These approaches are intrusive and inconvenient to use, especially for elderly people. Some non-intrusive vital signs monitoring approaches based on wireless sensing have been proposed in recent years. However, these approaches require the target user to be in situ during the monitoring process, which greatly limits their utilization in practical scenarios where the target users usually move around. In this paper, we propose RF-RMM, an RFID-based approach to accurate and continuous respiration monitoring for mobile users. The major challenge in respiration monitoring for moving people is that the tiny body displacement caused by the user's respiration is overwhelmed by the user's entire body movement. To address this issue, we propose a novel approach that uses a pair of tags to eliminate the effect of the user's body movement. We fuse the data from the paired tags to cancel the effect of the user's entire body movement and retain only the displacement caused by the user's respiration. Another challenging issue in implementing RF-RMM is how to resolve the phase ambiguity problem when the target user moves around, which becomes more serious than in the static case. We propose a distance tracking algorithm to track the phase transition during the user's movement, according to which the phase ambiguity problem can be well handled. We implement RF-RMM on commercial RFID devices and conduct extensive real-world experiments to evaluate its performance. The results show that RF-RMM achieves accurate respiration rate monitoring with an average error of 0.54 BPM in estimating different users' respiration rate and an average relative error of less than 13% in estimating the user's individual breath length.

In 1990, Hendry conjectured that all Hamiltonian chordal graphs are cycle extendable. After a series of papers confirming the conjecture for a number of graph classes, the conjecture is yet refuted by Lafond and Seamone in 2015. Given that their counterexamples are not strongly chordal graphs and they are all only 2-connected, Lafond and Seamone asked the following two questions: (1) Are Hamiltonian strongly chordal graphs cycle extendable? (2) Is there an integer $k$ such that all $k$-connected Hamiltonian chordal graphs are cycle extendable? Later, a conjecture stronger than Hendry's is proposed. In this paper, we resolve all these questions in the negative. On the positive side, we add to the list of cycle-extendable graphs two more graph classes, namely, Hamiltonian 4-fan-free chordal graphs, where every induced $K_5 - e$ has true twins, and Hamiltonian $\{4{\sc -fan}, \overline{A} \}$-free chordal graphs.

High-risk prediction of cardiovascular disease is of great significance and impendency in medical fields with the increasing phenomenon of sub-health these years. Most existing pathological methods for the prognosis prediction are either costly or prone to misjudgement. Therefore, plenty of automated models based on machine learning have been proposed to predict the onset of cardiovascular disease with the premorbid information of patients extracted from their historical Electronic Health Records (EHRs). However, it is a tough job to select proper features from longitudinal and heterogeneous EHRs, and also a great challenge to obtain accurate and robust representations for patients. In this paper, we propose an entirely end-to-end model called DeepRisk based on attention mechanism and deep neural networks, which can not only learn high-quality features automatically from EHRs, but also efficiently integrate heterogeneous and time-ordered medical data, and finally predict patients' risk of cardiovascular diseases. Experiments are carried out on a real medical dataset and results show that DeepRisk can significantly improve the high-risk prediction accuracy for cardiovascular disease compared with state-of-the-art approaches.

The dysregulation and mutation of long non-coding RNAs (lncRNAs) have been proved to result in a variety of human diseases. Identifying potential disease-related lncRNAs may benefit disease diagnosis, treatment and prognosis. A number of methods have been proposed to predict the potential lncRNA-disease relationships. However, most of them may give rise to incorrect results due to relying on single similarity measure. This article proposes a novel framework (ILDMSF) by fusing the lncRNA similarities and disease similarities, which are measured by lncRNA-related gene and known lncRNA-disease interaction and disease semantic interaction, and known lncRNA-disease interaction, respectively. Further, the support vector machine is employed to identify the potential lncRNA-disease associations based on the integrated similarity. The leave-one-out cross validation is performed to compare ILDMSF with other state of the art methods. The experimental results demonstrate our method is prospective in exploring potential correlations between lncRNA and disease.

Research of Protein-Protein Interaction (PPI) Network Alignment is playing an important role in understanding the crucial underlying biological knowledge such as functionally homologous proteins and conserved evolutionary pathways across different species. Existing methods of PPI network alignment often try to improve the coverage ratio of the alignment result by aligning all proteins from different species. However, there is a fundamental biological premise that needs to be considered carefully: not every protein in a species can, nor should, find its homologous proteins in other species. In this work, we propose a novel alignment method to map only those proteins with the most similarity throughout the PPI networks of multiple species. For the similarity features of the protein in the networks, we integrate both topological features with biological characteristics to provide enhanced supports for the alignment procedures. For topological features, we apply a representation learning method on the networks and generate a low dimensional vector embedding with its surrounding structural features for each protein. The topological similarity of proteins from different PPI networks can thus be transferred as the similarity of their corresponding vector representations, which provides a new way to comprehensively quantify the topological similarities between proteins. We also propose a new measure for the topological evaluation of the alignment results which better uncover the structural quality of the alignment across multiple networks. Both biological and topological evaluations on the alignment results of real datasets demonstrate our approach is promising and preferable against previous multiple alignment methods.

High-throughput screening technologies have provided a large amount of drug sensitivity data for a panel of cancer cell lines and hundreds of compounds. Computational approaches to analyzing these data can benefit anticancer therapeutics by identifying molecular genomic determinants of drug sensitivity and developing new anticancer drugs. In this study, we have developed a deep learning architecture to improve the performance of drug sensitivity prediction based on these data. We integrated both genomic features of cell lines and chemical information of compounds to predict the half maximal inhibitory concentrations [Formula: see text] on the Cancer Cell Line Encyclopedia (CCLE) and the Genomics of Drug Sensitivity in Cancer (GDSC) datasets using a deep neural network, which we called DeepDSC. Specifically, we first applied a stacked deep autoencoder to extract genomic features of cell lines from gene expression data, and then combined the compounds' chemical features to these genomic features to produce final response data. We conducted 10-fold cross-validation to demonstrate the performance of our deep model in terms of root-mean-square error (RMSE) and coefficient of determination [Formula: see text]. We show that our model outperforms the previous approaches with RMSE of 0.23 and [Formula: see text] of 0.78 on CCLE dataset, and RMSE of 0.52 and [Formula: see text] of 0.78 on GDSC dataset, respectively. Moreover, to demonstrate the prediction ability of our models on novel cell lines or novel compounds, we left cell lines originating from the same tissue and each compound out as the test sets, respectively, and the rest as training sets. The performance was comparable to other methods.

Breast cancer is a heterogeneous disease with many clinically distinguishable molecular subtypes each corresponding to a cluster of patients. Identification of prognostic and heterogeneous biomarkers for breast cancer is to detect cluster-specific gene biomarkers which can be used for accurate survival prediction of breast cancer outcomes. In this study, we proposed a FUsion Network-based method (FUNMarker) to identify prognostic and heterogeneous breast cancer biomarkers by considering the heterogeneity of patient samples and biological information from multiple sources. To reduce the affect of heterogeneity of patients, samples were first clustered using the K-means algorithm based on the principal components of gene expression. For each cluster, to comprehensively evaluate the influence of genes on breast cancer, genes were weighted from three aspects: biological function, prognostic ability and correlation with known disease genes. Then they were ranked via a label propagation model on a fusion network that combined physical protein interactions from seven types of networks and thus could reduce the impact of incompleteness of interactome. We compared FUNMarker with three state-of-the-art methods and the results showed that biomarkers identified by FUNMarker were biological interpretable and had stronger discriminative power than the existing methods in differentiating patients with different prognostic outcomes.

Assembling genomes from single-cell sequencing data is essential for single-cell studies. However, single-cell assemblies are challenging due to (i) the highly non-uniform read coverage and (ii) the elevated levels of sequencing errors and chimeric reads. Although several assemblers for single-cell data have been proposed in recent years, most of them fail to construct correct long contigs. In this study, we present a new framework called EPGA-SC for de novo assembly of single-cell sequencing reads. The EPGA assembler has designed strategies to solve the problems caused by sequencing errors, sequencing biases, and repetitive regions. However, the extremely unbalanced and richer error types prevent EPGA to achieve high performance in single-cell sequencing data. In this study, we designed EPGA-SC based on EPGA. The main innovations of EPGA-SC are as follows: (i) classifying reads to reduce the proportion of false reads; (ii) using multiple sets of high precision paired-end reads generated from the high precision assemblies produced by other assembler such as SPAdes to overcome the impact of sequencing biases and repetitive regions; and (iii) developing novel algorithms for removing chimeric errors and extending contigs. We test EPGA-SC with seven datasets. The experimental results show that EPGA-SC can generate better assemblies than most current tools in most time in term of MAX contig, N50, NG50, NA50, and NGA50.

Computational drug repositioning, which is an efficient approach to find potential indications for drugs, has been used to increase the efficiency of drug development. The drug repositioning problem essentially is a top-K recommendation task that recommends most likely diseases to drugs based on drug and disease related information. Therefore, many recommendation methods can be adopted to drug repositioning. Collaborative metric learning (CML) algorithm can produce distance metrics that capture the important relationships among objects, and has been widely used in recommendation domains. By applying CML in drug repositioning, a joint metric space is learned to encode drug's relationships with different diseases. In this study, we propose a novel drug repositioning computational method using Collaborative Metric Learning to predict novel drug-disease associations based on known drug and disease related information. Specifically, the proposed method learns latent vectors of drugs and diseases by applying metric learning, and then predicts the association probability of one drug-disease pair based on the learned vectors. The comprehensive experimental results show that CMLDR outperforms the other state-of-the-art drug repositioning algorithms in terms of precision, recall, and AUPR.

Single-cell RNA sequencing (scRNA-seq) technology provides quantitative gene expression profiles at single-cell resolution. As a result, researchers have established new ways to explore cell population heterogeneity and genetic variability of cells. One of the current research directions for scRNA-seq data is to identify different cell types accurately through unsupervised clustering methods. However, scRNA-seq data analysis is challenging because of their high noise level, high dimensionality and sparsity. Moreover, the impact of multiple latent factors on gene expression heterogeneity and on the ability to accurately identify cell types remains unclear. How to overcome these challenges to reveal the biological difference between cell types has become the key to analyze scRNA-seq data. For these reasons, the unsupervised learning for cell population discovery based on scRNA-seq data analysis has become an important research area. A cell similarity assessment method plays a significant role in cell clustering. Here, we present BioRank, a new cell similarity assessment method based on annotated gene sets and gene ranks. To evaluate the performances, we cluster cells by two classical clustering algorithms based on the similarity between cells obtained by BioRank. In addition, BioRank can be used by any clustering algorithm that requires a similarity matrix. Applying BioRank to 12 public scRNA-seq datasets, we show that it is better than or at least as well as several popular similarity assessment methods for single cell clustering.

With the development of high-through sequencing technology and microbiology, many studies have evidenced that microbes are associated with human diseases, such as obesity, liver cancer, and so on. Therefore, identifying the association between microbes and diseases has become an important study topic in current bioinformatics. The emergence of microbe-disease association database has provided an unprecedented opportunity to develop computational method for predicting microbe-disease associations. In the study, we propose a low-rank matrix completion method (called MCHMDA) to predict microbe-disease associations by integrating similarities of microbes and diseases and known microbe-disease associations into a heterogeneous network. The microbe similarity is computed from Gaussian Interaction Profile (GIP) kernel similarity based on the known microbe-disease associations. Then, we further improve the microbe similarity by taking into account the inhabiting organs of these microbes in human body. The disease similarity is computed by the average of disease GIP similarity, disease symptom-based similarity, and disease functional similarity. Then, we construct a heterogeneous microbe-disease association network by integrating the microbe similarity network, disease similarity network, and known microbe-disease association network. Finally, a matrix completion method is used to calculate the association scores of unknown microbe-disease pairs by the fast Singular Value Thresholding (SVT) algorithm. Via 5-fold Cross Validation (5CV) and Leave-One-Out Cross Validation (LOOCV), we evaluate the prediction performances of MCHMDA and other state-of-the-art methods which include BRWMDA, NGRHMDA, LRLSHMDA, and KATZHMDA. On benchmark dataset HMDAD, the experimental results show that MCHMDA outperforms other methods in terms of area under the receiver operating characteristic curve (AUC). MCHMDA achieves the AUC values of 0.9251 and 0.9495 in 5CV and LOOCV, respectively, which are the highest values among the competing methods. In addition, we also further indicate the prediction generality of MCHMDA on an expanded microbe-disease associations dataset (HMDAD-SUP). Finally, case studies prove the prediction ability in practical applications.

Computational methods including centrality and machine learning-based methods have been proposed to identify essential proteins for understanding the minimum requirements of the survival and evolution of a cell. In centrality methods, researchers are required to design a score function which is based on prior knowledge, yet is usually not sufficient to capture the complexity of biological information. In machine learning-based methods, some selected biological features cannot represent the complete properties of biological information as they lack a computational framework to automatically select features. To tackle these problems, we propose a deep learning framework to automatically learn biological features without prior knowledge. We use node2vec technique to automatically learn a richer representation of protein-protein interaction (PPI) network topologies than a score function. Bidirectional long short term memory cells are applied to capture non-local relationships in gene expression data. For subcellular localization information, we exploit a high dimensional indicator vector to characterize their feature. To evaluate the performance of our method, we tested it on PPI network of S. cerevisiae. Our experimental results demonstrate that the performance of our method is better than traditional centrality methods and is superior to existing machine learning-based methods. To explore which of the three types of biological information is the most vital element, we conduct an ablation study by removing each component in turn. Our results show that the PPI network embedding contributes most to the improvement. In addition, gene expression profiles and subcellular localization information are also helpful to improve the performance in identification of essential proteins.

A growing amount of evidence suggests that long non-coding RNAs (lncRNAs) play important roles in the regulation of biological processes in many human diseases. However, the number of experimentally verified lncRNA-disease associations is very limited. Thus, various computational approaches are proposed to predict lncRNA-disease associations. Current matrix factorization-based methods cannot capture the complex non-linear relationship between lncRNAs and diseases, and traditional machine learning-based methods are not sufficiently powerful to learn the representation of lncRNAs and diseases. Considering these limitations in existing computational methods, we propose a deep matrix factorization model to predict lncRNA-disease associations (DMFLDA in short). DMFLDA uses a cascade of non-linear hidden layers to learn latent representation to represent lncRNAs and diseases. By using non-linear hidden layers, DMFLDA captures the more complex non-linear relationship between lncRNAs and diseases than traditional matrix factorization-based methods. In addition, DMFLDA learns features directly from the lncRNA-disease interaction matrix and thus can obtain more accurate representation learning for lncRNAs and diseases than traditional machine learning methods. The low dimensional representations of the lncRNAs and diseases are fused to estimate the new interaction value. To evaluate the performance of DMFLDA, we perform leave-one-out cross-validation and 5-fold cross-validation on known experimentally verified lncRNA-disease associations. The experimental results show that DMFLDA performs better than the existing methods. The case studies show that many predicted interactions of colorectal cancer, prostate cancer, and renal cancer have been verified by recent biomedical literature. The source code and datasets can be obtained from https://github.com/CSUBioGroup/DMFLDA.

Homozygous and heterozygous deletions commonly exist in the human genome. For current structural variation detection tools, it is significant to determine whether a deletion is homozygous or heterozygous. However, the problems of sequencing errors, micro-homologies, and micro-insertions prohibit common alignment tools from identifying accurate breakpoint locations, and often result in detecting false structural variations. In this study, we present a novel deletion detection tool called Sprites2. Comparing with Sprites, Sprites2 makes the following modifications: (1) The distribution of insert size is used in Sprites2, which can identify the type of deletions and improve the accuracy of deletion calls. (2) A precise alignment method based on AGE (one algorithm simultaneously aligning 5' and 3' ends between two sequences) is adopted in Sprites2 to identify breakpoints, which is helpful to resolve the problems introduced by sequencing errors, micro-homologies, and micro-insertions. In order to test and verify the performance of Sprites2, some simulated and real datasets are adopted in our experiments, and Sprites2 is compared with five popular tools. The experimental results show that Sprites2 can improve the performance of deletion detection. Sprites2 can be downloaded from https://github.com/zhangzhen/sprites2.

Gene regulatory networks (GRNs) play a key role in biological processes. However, GRNs are diverse under different biological conditions. Reconstructing gene regulatory networks (GRNs) from gene expression has become an important opportunity and challenge in the past decades. Although there are a lot of existing methods to infer the topology of GRNs, such as mutual information, random forest, and partial least squares, the accuracy is still low due to the noise and high dimension of the expression data. In this paper, we introduce an ensemble Multivariate Adaptive Regression Splines (MARS) based method to reconstruct the directed GRNs from multifactorial gene expression data, called PBMarsNet. PBMarsNet incorporates part mutual information (PMI) to pre-weight the candidate regulatory genes and then uses MARS to detect the nonlinear regulatory links. Moreover, we apply bootstrap to run the MARS multiple times and average the outputs of each MARS as the final score of regulatory links. The results on DREAM4 challenge and DREAM5 challenge datasets show PBMarsNet has a superior performance and generalization over other state-of-the-art methods.

Data center networks typically adopt multi-rooted tree topologies to provide high bisection bandwidth. Various fine-grained load balancing schemes have been proposed to split flows across multiple paths. However, data center networks suffer from many uncertainties such as highly dynamic traffic. These uncertainties easily make network become asymmetric, resulting in significant packet reordering. Unfortunately, existing solutions passively deal with packet reordering based on a threshold and hardly adapt to asymmetric networks because of lacking the explicit reordering feedback. These solutions either fail to quickly respond to packet loss or cause unnecessary fast retransmission, which reduces link utilization and increases flow completion time. In this paper, we propose a fine-grained load balancing scheme RMC to eliminate the impact of packet reordering and handle uncertainties in asymmetric networks. To avoid unnecessary fast retransmission, the switch proactively identifies reordered packet according to local queue length and global path latency. Furthermore, we employ a coding technique with redundancy optimization to reduce long-tailed flow completion time under network asymmetry. Through a series of large-scale NS2 simulations and testbed experiments, we demonstrate that RMC effectively avoids unnecessary fast retransmission under different network scenarios and reduces flow completion time by up to 72% compared with state-of-the-art schemes.

In recent years, more and more evidence indicates that circular RNAs (circRNAs) with covalently closed loop play various roles in biological processes. Dysregulation and mutation of circRNAs may be implicated in diseases. Due to its stable structure and resistance to degradation, circRNAs provide great potential to be diagnostic biomarkers. Therefore, predicting circRNA-disease associations is helpful in disease diagnosis. However, there are few experimentally validated associations between circRNAs and diseases. Although several computational methods have been proposed, precisely representing underlying features and grasping the complex structures of data are still challenging. In this paper, we design a new method, called DMFCDA (Deep Matrix Factorization CircRNA-Disease Association), to infer potential circRNA-disease associations. DMFCDA takes both explicit and implicit feedback into account. Then, it uses a projection layer to automatically learn latent representations of circRNAs and diseases. With multi-layer neural networks, DMFCDA can model the non-linear associations to grasp the complex structure of data. We assess the performance of DMFCDA using leave-one cross-validation and 5-fold cross-validation on two datasets. Computational results show that DMFCDA efficiently infers circRNA-disease associations according to AUC values, the percentage of precisely retrieved associations in various top ranks, and statistical comparison. We also conduct case studies to evaluate DMFCDA. All results show that DMFCDA provides accurate predictions.

The state-of-the-art datacenter load balancing designs commonly optimize bisection bandwidth with homogeneous switching granularity. Their performances surprisingly degrade under mixed traffic containing both short and long flows. Specifically, the short flows suffer from long-tailed delay, while the throughputs of long flows also degrade dramatically due to low link utilization and packet reordering. To solve these problems, we design a traffic-aware load balancing (TLB) scheme to adaptively adjust the switching granularity of long flows according to the load strength of short ones. Under the heavy load of short flows, the long flows use large switching granularity to help short ones obtain more opportunities in choosing short queues to complete quickly. On the contrary, the long flows reroute flexibly with small switching granularity to achieve high throughput. Furthermore, under extremely bursty scenario, we utilize the packet slicing scheme for long flows to release bandwidth for short ones. The experimental results of NS2 simulation and testbed implementation show that TLB significantly reduces the average flow completion time of short flows by 16%-67% over the state-of-the-art load balancers and achieves the high throughput for long flows. Moreover, for extreme bursty case, at the acceptable throughput degradation of long flows, TLB with packet slicing reduces the deadline missing ratio of bursty short flows by up to 80%.

Modern data center networks are usually constructed in multi-rooted tree topologies, which require the highly efficient multi-path load balancing to achieve high link utilization. Recent packet-level load balancer obtains high throughput by spraying packets to all paths, but it easily leads to the packet reordering under network asymmetry. The flow-level or flowlet-level load balancer avoids the packet reordering, while reducing the link utilization due to their inflexibility. To solve these problems, we design a Queueing Delay Aware Packet Spraying (QDAPS), that effectively mitigates the packet reordering for packet-level load balancer. QDAPS selects paths for packets according to the queueing delay of output buffer, and lets the packet arriving earlier be forwarded before the later packets to avoid packet reordering. Moreover, we adopt the “power-of- n-choices” paradigm on QDAPS to alleviate the impact of herd behavior under multiple forwarding engines. We compare QDAPS with ECMP, LetFlow and RPS through NS2 simulation and Mininet implementation. The test results show that QDAPS reduces flow completion time (FCT) by ~30%-50% over the state-of-the-art load balancing mechanism.

In data center networks, many network-intensive applications leverage large fan-in and many-to-one communication to achieve high performance. However, the special traffic patterns, such as micro-burst and high concurrency, easily cause TCP Incast problem and seriously degrade the application performance. To address the TCP Incast problem, we first reveal theoretically and empirically that alleviating packet burstiness is much more effective in reducing the Incast probability than controlling the congestion window. Inspired by the findings and insights from our experimental observations, we further propose a general supporting scheme Adaptive Pacing (AP), which dynamically adjusts burstiness according to the flow concurrency without any change on switch. Additionally, a sender-based approach is proposed to estimate the flow concurrency. Another feature of AP is its broad applicability. We integrate AP transparently into different TCP protocols (i.e., DCTCP, L2DCT and D2TCP). Through a series of large-scale NS2 simulations and testbed experiments, we show that AP significantly reduces the Incast probability across different TCP protocols and the network goodput can be increased consistently by on average 7× under severe congestion.

To identify and explore the similarities between diseases is of great significance for u nderstanding t he pathogenic mechanisms of emerging complex diseases. Some methods try to measure the similarity of diseases through deep learning models. However, the insufficient number of labelled similar disease pairs cannot support the optimal training of the models. In this paper, we propose a Multi-Task Graph Neural Network (MTGNN) framework to retrieve similar diseases by few-shot learning. To deal with the problem of insufficient number o f labelled similar disease pairs, we design double tasks to optimize the graph neural network for disease similarity task (lack of labelled training data) by introducing link prediction task (sufficient labelled training data). The similarity between diseases can then be obtained by measuring the distance between disease embeddings in high-dimensional space learning from the double tasks. The experiment results illustrate the overall effectiveness by comparing with prior methods on few labeled training dataset.

The cause of Alzheimer's disease (AD) is insufficient to understand so far, and its diagnosis is challenging in clinical practice. Recently, the convolutional neural network (CNN) model has shown impressive performance in medical image analysis. Combining CNN with magnetic resonance imaging (MRI) image has excellent potential for AD diagnosis. However, it is still a challenging task. To address the challenge, we propose a joint learning method based on multi-scale representation (JL-MSR). The multi-scale representation is proposed to obtain more feature maps by the multi-scale atrous convolutions. Furthermore, in order to use the intrinsic relationship between diagnostic results and clinical scores, we propose a joint learning strategy using the diagnosis result as the primary label and the Mini-Mental State Examination (MMSE) score as the secondary label to joint training. The proposed method is evaluated on a dataset of 417 subjects (including 188 AD and 229 health controls (HC)) from the Alzheimer's Disease Neuroimaging Initiative (ADNI). The experimental results show that our proposed method achieves an accuracy of 88.1% and an area under the receiver operating characteristic (ROC) curve (AUC) value of 0.942 for AD diagnosis, respectively. Compared with a state-of-the-art method in AD diagnosis, our proposed method performs better, and has potential in clinical diagnosis.

Medical image segmentation is the primary measure of medical image analysis. With the development of deep learning, U-net based approaches have presented for different medical image segmentation tasks. However, the pooling and the simple convolution operation for deep feature maps in the U-shaped network would lead to the coarse segmentation result. In this paper, we design a local adaptive U-net (LA U-net) for medical image segmentation. There are two major modules: the Local Adaptive Module (LAM) and Multi-scale Convolution Module (MCM) in the network. The LAM get more feature maps from each down-sampling process. The MCM capture more global information for the encoding path. To validate the proposed network's performance, we verify it on two datasets: DRIVE dataset, and ISIC 2018 dataset; the results show that LA U-net achieves superior performance on two datasets.

Correctly identifying the potential Anatomical Therapeutic Chemical (ATC) codes for drugs can accelerate drug development and reduce the cost of experiments. However, most of the existing methods only analyze the first-level ATC code of drugs and lack of the ability to learn basic features from sparsely known drug-ATC code associations. In this paper, we propose a novel method based on deep residual network framework, named RNPredATC, to predict potential drug-ATC code associations by integrating drug structure similarity, ATC sematic similarity, and known drug-ATC code associations. RNPredATC can extract dense feature vectors from sparsely known drug-ATC code associations and reduce the impact from degradation problem, such as gradient vanishing or gradient explosion of deep network. The experimental results show that RNPredATC achieves better performances.

Recommending stock with the highest return ratio is always a challenging problem in the field of financial technology. In this paper, we propose a time-aware graph relational attention network (TRAN) for stock recommendation based on return ratio ranking. In TRAN, time-aware relational attention mechanism is the key unit to capture time-varying correlation strength between stocks by the interaction of historical sequences and stock description documents. With the dynamic strength, the nodes of the stock relation graph aggregate the features of neighbor stock nodes by graph convolution operation. For a given group of stocks, our model can output the ranking results of stocks according to their return ratios. The experimental results on several real-world datasets demonstrate the effectiveness of our TRAN for stock recommendation.

To deliver video in a high quality across various network conditions, adaptive bitrate (ABR) algorithms dynamically select bitrate for each chunk according to perceived network rate and buffer occupancy. Unfortunately, though ameliorating the quality of chunks with dynamic scenes can obtain more QoE gain than the ones with static scenes, current ABR algorithms generally aim to maximize the average bitrate rather than perceptual quality, resulting in the QoE degradation. To address this issue, we propose a dynamic-chunk quality aware adaptive bitrate scheme via apprenticeship learning named DAVS, in which higher quality is chosen for the dynamic chunks without decreasing the quality of static chunks excessively. The experimental results show that DAVS enhances the quality of dynamic chunks and greatly improves the overall QoE compared with the state-of-the-art ABR algorithms.

Fractional repetition (FR) codes are a class of repair efficient erasure codes that can recover a failed storage node with both optimal repair bandwidth and complexity. In this paper, we focus on the minimum distance of FR codes, which is the smallest number of nodes whose failure leads to the unrecoverable loss of stored files. We consider upper bounds on the minimum distance and present several families of FR codes attaining these bounds. The optimal constructions are derived from regular graphs and combinatorial designs, respectively.

Since most flows are short-lived in data center networks, fast convergence becomes very important to help the short flows effectively utilize high bandwidth. Though current feedback-based transport control protocols (TCPs) provide fast convergence via fine-grained explicit congestion information from customized switches, they unavoidably incur large traffic overhead for widely existing small packets in data center applications, resulting in suboptimal network efficiency. To solve this issue, we propose a datacenter TCP based on differential feedbacks, called DECN, to achieve fast convergence without any traffic overhead. Specifically, DECN feeds rate difference between the target and current rate back to the source by using multiple consecutive packets. The experimental results of NS2 simulation and testbed implementation show that DECN achieves comparable fast convergence as XCP without incurring any extra feedback overhead. Compared with the state-of-the-art feedback-based TCPs, DECN reduces the flow completion time by up to 34.1% in typical data center applications.

In distributed key-value stores, multiple replica servers are always available for each key-value access operation when the eventually consistency model is employed. Accordingly, the replica selection algorithm is crucial to the tail latency of the key-value access operations generated by end-user requests, especially under the environment of heterogeneous replica servers. The main challenge of making replica selection decision is to know the status of replica serves with acceptable overhead for each lightweight key-value access operation. Recently, aware of the time-varying performance and load of replica servers, the adaptive replica selection algorithm C3 suggests piggybacking the information of replica server via the returned value” to guide the replica selection. Although the good performance of C3 has been verified by experiments and simulations, the poor timeliness of feedback information makes a large performance gap between C3 and the ideal replica selection algorithm. To narrow this gap, we propose the information dissemination mechanism for C3, which lets both “client” and “server” store the records about the status of replica servers, and both key” and value” piggyback multiple records. In this way, the information dissemination bottleneck at “slow” server would be removed with the help of multiple clients and servers. As confirmed by simulation results, the information dissemination mechanism can significantly improve the timeliness of feedback information with acceptable overhead and accordingly helps C3 to greatly reduce the tail latency (by about 35% under the default scenario).

To deliver high quality video across different network conditions, the video chunks are explicitly fetched by client or proactively pushed by server in Dynamic Adaptive Streaming over HTTP (DASH) system. Unfortunately, on the one hand, the client fetch mechanism suffers from bandwidth wastage due to its stop-and-wait fashion when the network delay becomes large. On the other hand, the server push mechanism performs poorly because of its inflexibility in bitrate switching under fluctuating bandwidth. To address these inefficiencies, we propose a pipeline-based chunk scheduling scheme called PCS to auto-turn the sending time of each chunk. For a given ABR algorithm, PCS dynamically pre-schedules the chunk delivery according to the real-time network conditions. Using the pipelined-based chunk delivery, PCS flexibly adjusts the bitrate of each chunk and meanwhile avoids the unnecessary waiting time in the stop-and-wait transmission. The experimental results of testbed implementations show that PCS greatly improves the average bitrate of the state-of-the-art ABR algorithms by up to 26%, and reduces the rebuffer rate by up to 31%.

Modern data centers often host multiple applications with diverse network demands. To provide fair bandwidth allocation to several thousand traversing flows, Approximate Fair Queueing (AFQ) utilizes multiple priority queues in switch to approximate ideal fair queueing. However, due to limited number of queues in commodity switches, AFQ easily experiences high packet loss and low link utilization. In this paper, we propose Elastic Fair Queueing (EFQ), which leverages limited priority queues to flexibly achieve both high network utilization and fair bandwidth allocation. EFQ dynamically assigns the free buffer space in priority queues for each packet to obtain high utilization without sacrificing flow-level fairness. The results of simulation experiments and real implementations show that EFQ reduces the average flow completion time by up to 82% over the state-of-the-art fair bandwidth allocation mechanisms.

Cloud applications generate a variety of workloads ranging from delay-sensitive flows to bandwidth-hungry ones in data centers. Existing reactive or proactive congestion control protocols are hard to simultaneously achieve ultra-low latency and high link utilization across all workloads in data center networks. We present a new receiver-driven transport scheme using anti-ECN (Explicit Congestion Notification) marking to achieve both near-zero queueing delay and full link utilization by reasonably increasing sending rate in the case of under-utilization. Specifically, switches mark the ECN bit of data packets once detecting spare bandwidth. When receiving the anti-ECN marked packet, the receiver generates the corresponding marked grant to trigger more data packets. The experimental results of small-scale testbed implementation and large-scale NS2 simulation show that AMRT effectively reduces the average flow completion time (AFCT) by up to 40.8% and improves the link utilization by up to 36.8% under high workload over the state-of-the-art receiver-driven transmission schemes.

The 40-100Gbps Energy Efficient Ethernet (EEE) standardized by IEEE 802.3bj contains not only the DeepSleep mode, which is 10% of the normal power consumption, but also the FastWake mode, which has much less state transition time and 70% of the normal power consumption. Correspondingly, the strategy of determining the state transitions is crucial to both the power consumption and the incurred latency of frames. As EEE applications always place tail latency constraint on frame transmission, existing strategies for the 40-100Gbps EEE are rarely configured with proper parameters for good power saving under the given tail latency constraint, especially in face of the variable traffic load. To address this problem, we design the Periodic Strategy (PS) for the 40-100Gbps EEE. Specifically, PS works periodically and optimizes the power saving independently in each cycle. In this way, PS decouples the latency constraint and energy saving. 1) It limits the largest incurred latency by the length of the cycle to satisfy the given latency constraint. 2) It achieves good power saving by selecting the proper low power state based on the prediction of incoming frames and entering the selected state just once in each cycle. What’s more, the good performance of PS is insensitive to the traffic pattern. Extensive simulations confirm that PS achieves better energy saving than all the existing strategies under the given latency constraint, under several kinds of different traffic patterns.

Recently, numerous novel transport protocols are proposed for the low latency of applications deployed in data center networks, e.g., web search and retail recommendation system. The state-of-art receiver-driven protocols, e.g., Homa and NDP, show the superior performance for achieving the lowest possible latency. However, Homa assumes that the core layer in data center network has no congestion, which limits its application for the existing over-subscribed networks. NDP requires to modify the switch hardware since it trims packets to headers when the packets cause the switch buffer to overflow, resulting in high deployment cost. In this paper, we present Polo to realize low latency for flows over commodity network fabric relying on Explicit Congestion Notification (ECN) and priority queues. According to packets with ECN marking, the Polo receiver obtains the congestion information such that it dynamically adjusts the number of data packets in network for maintaining the small switch queue. The adjustment is carried out periodically. The time interval is determined by keeping the extra high priority packet always in flight nor by a fine-grained timer. Further, Polo designs the packet recovery mechanisms to retransmit the lost packets as soon as possible. Simulation experiment results show that Polo outperforms the state-of-art receiver-driven protocols in a wide range of scenarios including incast.

In this paper, we present a framework for designing FPT approximation algorithms for many k-clustering problems. Our results are based on a new technique for reducing search spaces. A reduced search space is a small subset of the input data that has the guarantee of containing k clients close to the facilities opened in an optimal solution for any clustering problem we consider. We show, somewhat surprisingly, that greedily sampling O(k) clients yields the desired reduced search space, based on which we obtain FPT(k)-time algorithms with improved approximation guarantees for problems such as capacitated clustering, lower-bounded clustering, clustering with service installation costs, fault tolerant clustering, and priority clustering.

Tag searching is a fundamental problem for a variety of radio frequency identification (RFID) applications. Prior works focus on single group searching, which refers to determining which ones in a given set of tags exist in the system. In this paper, we propose PTS, a protocol that can perform fast Parallel Tag Searching for multiple users simultaneously. Different from prior works that have to execute k times to search k groups separately, PTS obtains searching results for all the k groups with only one-shot execution. PTS achieves high parallelism due to some novel designs. First, we develop a grouping filter that encodes the membership of tags in different groups, with which non-target tags can be efficiently filtered out for all the k groups simultaneously. Second, we design two new codes, grouping code and mapping code, with which the remaining tags can quickly verify and confirm which group they belong to. We theoretically analyze how to set optimal parameters for PTS to minimize the execution time and conduct extensive simulation experiments to evaluate its performance. Compared with the state-of-the-art solutions, PTS significantly improves time efficiency in multiple group searching scenarios (by a factor of up to 7.54X when k = 10) and achieves the same time efficiency in single group searching scenarios.

Predicting the survival of cancer patients holds significant meaning for public health, and has attracted increasing attention in medical information communities. In this study, we propose a novel framework for cancer survival prediction named Multimodal Graph Neural Network (MGNN), which explores the features of real-world multimodual data such as gene expression, copy number alteration and clinical data in a unified framework. In order to explore the inherent relation, we first construct the bipartite graphs between patients and multimodal data. Subsequently, graph neural network is adopted to obtain the embedding of each patient on different bipartite graphs. Finally, a multimodal fusion neural layer is designed to fuse the features from different modal data. The output of our method is the classification of short term survival or long term survival for each patient. Experimental results on one breast cancer dataset demonstrate that MGNN outperforms all baselines. Furthermore, we test the trained model on lung cancer dataset, and the experimental results verify the strong robust by comparing with state-of-the-art methods.

SUMMARY: The recent advance in genome engineering technologies based on CRISPR/Cas9 system is enabling people to systematically understand genomic functions. A short RNA string (the CRISPR guide RNA) can guide the Cas9 endonuclease to specific locations in complex genomes to cut DNA double-strands. The CRISPR guide RNA is essential for gene editing systems. Recently, the GuideScan software is developed to design CRISPR guide RNA libraries, which can be used for genome editing of coding and non-coding genomic regions effectively. However, GuideScan is a serial program and computationally expensive for designing CRISPR guide RNA libraries from large genomes. Here, we present an efficient guide RNA library designing tool (MultiGuideScan) by implementing multiple processes of GuideScan. MultiGuideScan speeds up the guide RNA library designing about 9-12 times on a 32-process mode comparing to GuideScan. MultiGuideScan makes it possible to design guide RNA libraries from large genomes. AVAILABILITY AND IMPLEMENTATION: MULTIGUIDESCAN IS AVAILABLE AT GITHUB: https://github.com/bioinfomaticsCSU/MultiGuideScan. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: Determining the structures of proteins is a critical step to understand their biological functions. Crystallography-based X-ray diffraction technique is the main method for experimental protein structure determination. However, the underlying crystallization process, which needs multiple time-consuming and costly experimental steps, has a high attrition rate. To overcome this issue, a series of in silico methods have been developed with the primary aim of selecting the protein sequences that are promising to be crystallized. However, the predictive performance of the current methods is modest. RESULTS: We propose a deep learning model, so-called CLPred, which uses a bidirectional recurrent neural network with long short-term memory (BLSTM) to capture the long-range interaction patterns between k-mers amino acids to predict protein crystallizability. Using sequence only information, CLPred outperforms the existing deep-learning predictors and a vast majority of sequence-based diffraction-quality crystals predictors on three independent test sets. The results highlight the effectiveness of BLSTM in capturing non-local, long-range inter-peptide interaction patterns to distinguish proteins that can result in diffraction-quality crystals from those that cannot. CLPred has been steadily improved over the previous window-based neural networks, which is able to predict crystallization propensity with high accuracy. CLPred can also be improved significantly if it incorporates additional features from pre-extracted evolutional, structural and physicochemical characteristics. The correctness of CLPred predictions is further validated by the case studies of Sox transcription factor family member proteins and Zika virus non-structural proteins. AVAILABILITY AND IMPLEMENTATION: https://github.com/xuanwenjing/CLPred.

MOTIVATION: Protein-protein interactions (PPIs) play important roles in many biological processes. Conventional biological experiments for identifying PPI sites are costly and time-consuming. Thus, many computational approaches have been proposed to predict PPI sites. Existing computational methods usually use local contextual features to predict PPI sites. Actually, global features of protein sequences are critical for PPI site prediction. RESULTS: A new end-to-end deep learning framework, named DeepPPISP, through combining local contextual and global sequence features, is proposed for PPI site prediction. For local contextual features, we use a sliding window to capture features of neighbors of a target amino acid as in previous studies. For global sequence features, a text convolutional neural network is applied to extract features from the whole protein sequence. Then the local contextual and global sequence features are combined to predict PPI sites. By integrating local contextual and global sequence features, DeepPPISP achieves the state-of-the-art performance, which is better than the other competing methods. In order to investigate if global sequence features are helpful in our deep learning model, we remove or change some components in DeepPPISP. Detailed analyses show that global sequence features play important roles in DeepPPISP. AVAILABILITY AND IMPLEMENTATION: The DeepPPISP web server is available at http://bioinformatics.csu.edu.cn/PPISP/. The source code can be obtained from https://github.com/CSUBioGroup/DeepPPISP. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

BACKGROUND: One of the most essential problems in structural bioinformatics is protein fold recognition. In this paper, we design a novel deep learning architecture, so-called DeepFrag-k, which identifies fold discriminative features at fragment level to improve the accuracy of protein fold recognition. DeepFrag-k is composed of two stages: the first stage employs a multi-modal Deep Belief Network (DBN) to predict the potential structural fragments given a sequence, represented as a fragment vector, and then the second stage uses a deep convolutional neural network (CNN) to classify the fragment vector into the corresponding fold. RESULTS: Our results show that DeepFrag-k yields 92.98% accuracy in predicting the top-100 most popular fragments, which can be used to generate discriminative fragment feature vectors to improve protein fold recognition. CONCLUSIONS: There is a set of fragments that can serve as structural "keywords" distinguishing between major protein folds. The deep learning architecture in DeepFrag-k is able to accurately identify these fragments as structure features to improve protein fold recognition.

BACKGROUND: Repetitive sequences account for a large proportion of eukaryotes genomes. Identification of repetitive sequences plays a significant role in many applications, such as structural variation detection and genome assembly. Many existing de novo repeat identification pipelines or tools make use of assembly of the high-frequency k-mers to obtain repeats. However, a certain degree of sequence coverage is required for assemblers to get the desired assemblies. On the other hand, assemblers cut the reads into shorter k-mers for assembly, which may destroy the structure of the repetitive regions. For the above reasons, it is difficult to obtain complete and accurate repetitive regions in the genome by using existing tools. RESULTS: In this study, we present a new method called RepAHR for de novo repeat identification by assembly of the high-frequency reads. Firstly, RepAHR scans next-generation sequencing (NGS) reads to find the high-frequency k-mers. Secondly, RepAHR filters the high-frequency reads from whole NGS reads according to certain rules based on the high-frequency k-mer. Finally, the high-frequency reads are assembled to generate repeats by using SPAdes, which is considered as an outstanding genome assembler with NGS sequences. CONLUSIONS: We test RepAHR on five data sets, and the experimental results show that RepAHR outperforms RepARK and REPdenovo for detecting repeats in terms of N50, reference alignment ratio, coverage ratio of reference, mask ratio of Repbase and some other metrics.

BACKGROUND: The identification of early mild cognitive impairment (EMCI), which is an early stage of Alzheimer's disease (AD) and is associated with brain structural and functional changes, is still a challenging task. Recent studies show great promises for improving the performance of EMCI identification by combining multiple structural and functional features, such as grey matter volume and shortest path length. However, extracting which features and how to combine multiple features to improve the performance of EMCI identification have always been a challenging problem. To address this problem, in this study we propose a new EMCI identification framework using multi-modal data and graph convolutional networks (GCNs). Firstly, we extract grey matter volume and shortest path length of each brain region based on automated anatomical labeling (AAL) atlas as feature representation from T1w MRI and rs-fMRI data of each subject, respectively. Then, in order to obtain features that are more helpful in identifying EMCI, a common multi-task feature selection method is applied. Afterwards, we construct a non-fully labelled subject graph using imaging and non-imaging phenotypic measures of each subject. Finally, a GCN model is adopted to perform the EMCI identification task. RESULTS: Our proposed EMCI identification method is evaluated on 210 subjects, including 105 subjects with EMCI and 105 normal controls (NCs), with both T1w MRI and rs-fMRI data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database. Experimental results show that our proposed framework achieves an accuracy of 84.1% and an area under the receiver operating characteristic (ROC) curve (AUC) of 0.856 for EMCI/NC classification. In addition, by comparison, the accuracy and AUC values of our proposed framework are better than those of some existing methods in EMCI identification. CONCLUSION: Our proposed EMCI identification framework is effective and promising for automatic diagnosis of EMCI in clinical practice.

BACKGROUND: Cervical cancer is the fourth most common tumor in women worldwide, mostly resulting from high-risk human papillomavirus (HR-HPV) with persistent infection. RESULTS: The present discoveries are comprised of the following: (i) A total of 16.64% of the individuals were positive for HR-HPV infection, with 13.04% having a single HR-HPV type and 3.60% having multiple HR-HPV types. (ii) Cluster analysis showed that the infection rate trends of HPV31 and HPV33 in all infections as well as HPV33 and HPV35 in single infections in precancerous stages were very similar. (iii) The single/multiple infection proportions of HR-HPV demonstrated a trend that the multiple infections rates of HR-HPV increased as the disease developed. CONCLUSIONS: The HR-HPV prevalence in outpatients was 16.64%, and the predominant HR-HPV types in the study were HPV52, HPV58 and HPV16. HR-HPV subtypes with common biological properties had similar infection rate trends in precancerous stages. Especially, as the disease development of precancer evolved, defense against HPV infection broke, meanwhile, the potential of more HPV infection increased, which resulted in increase of multiple infections of HPV.

BACKGROUND: MicroRNAs (miRNAs) are a kind of small noncoding RNA molecules that are direct posttranscriptional regulations of mRNA targets. Studies have indicated that miRNAs play key roles in complex diseases by taking part in many biological processes, such as cell growth, cell death and so on. Therefore, in order to improve the effectiveness of disease diagnosis and treatment, it is appealing to develop advanced computational methods for predicting the essentiality of miRNAs. RESULT: In this study, we propose a method (PESM) to predict the miRNA essentiality based on gradient boosting machines and miRNA sequences. First, PESM extracts the sequence and structural features of miRNAs. Then it uses gradient boosting machines to predict the essentiality of miRNAs. We conduct the 5-fold cross-validation to assess the prediction performance of our method. The area under the receiver operating characteristic curve (AUC), F-measure and accuracy (ACC) are used as the metrics to evaluate the prediction performance. We also compare PESM with other three competing methods which include miES, Gaussian Naive Bayes and Support Vector Machine. CONCLUSION: The results of experiments show that PESM achieves the better prediction performance (AUC: 0.9117, F-measure: 0.8572, ACC: 0.8516) than other three computing methods. In addition, the relative importance of all features also further shows that newly added features can be helpful to improve the prediction performance of methods.

microRNAs (miRNAs) are small and important non-coding RNAs that regulate gene expression in transcriptional and post-transcriptional level by combining with their targets (genes). Predicting miRNA targets is an important problem in biological research. It is expensive and time-consuming to identify miRNA targets by using biological experiments. Many computational methods have been proposed to predict miRNA targets. In this study, we develop a novel method, named miRTRS, for predicting miRNA targets based on a recommendation algorithm. miRTRS can predict targets for an isolated (new) miRNA with miRNA sequence similarity, as well as isolated (new) targets for a miRNA with gene sequence similarity. Furthermore, when compared to supervised machine learning methods, miRTRS does not need to select negative samples. We use 10-fold cross validation and independent datasets to evaluate the performance of our method. We compared miRTRS with two most recently published methods for miRNA target prediction. The experimental results have shown that our method miRTRS outperforms competing prediction methods in terms of AUC and other evaluation metrics.

Identifying essential proteins plays an important role in disease study, drug design, and understanding the minimal requirement for cellular life. Computational methods for essential proteins discovery overcome the disadvantages of biological experimental methods that are often time-consuming, expensive, and inefficient. The topological features of protein-protein interaction (PPI) networks are often used to design computational prediction methods, such as Degree Centrality (DC), Betweenness Centrality (BC), Closeness Centrality (CC), Subgraph Centrality (SC), Eigenvector Centrality (EC), Information Centrality (IC), and Neighborhood Centrality (NC). However, the prediction accuracies of these individual methods still have space to be improved. Studies show that additional information, such as orthologous relations, helps discover essential proteins. Many researchers have proposed different methods by combining multiple information sources to gain improvement of prediction accuracy. In this study, we find that essential proteins appear in triangular structure in PPI network significantly more often than nonessential ones. Based on this phenomenon, we propose a novel pure centrality measure, so-called Neighborhood Closeness Centrality (NCC). Accordingly, we develop a new combination model, Extended Pareto Optimality Consensus model, named EPOC, to fuse NCC and Orthology information and a novel essential proteins identification method, NCCO, is fully proposed. Compared with seven existing classic centrality methods (DC, BC, IC, CC, SC, EC, and NC) and three consensus methods (PeC, ION, and CSC), our results on S.cerevisiae and E.coli datasets show that NCCO has clear advantages. As a consensus method, EPOC also yields better performance than the random walk model.

The rapid development of proteomics and high-throughput technologies has produced a large amount of Protein-Protein Interaction (PPI) data, which makes it possible for considering dynamic properties of protein interaction networks (PINs) instead of static properties. Identification of protein complexes from dynamic PINs becomes a vital scientific problem for understanding cellular life in the post genome era. Up to now, plenty of models or methods have been proposed for the construction of dynamic PINs to identify protein complexes. However, most of the constructed dynamic PINs just focus on the temporal dynamic information and thus overlook the spatial dynamic information of the complex biological systems. To address the limitation of the existing dynamic PIN analysis approaches, in this paper, we propose a new model-based scheme for the construction of the Spatial and Temporal Active Protein Interaction Network (ST-APIN) by integrating time-course gene expression data and subcellular location information. To evaluate the efficiency of ST-APIN, the commonly used classical clustering algorithm MCL is adopted to identify protein complexes from ST-APIN and the other three dynamic PINs, NF-APIN, DPIN, and TC-PIN. The experimental results show that, the performance of MCL on ST-APIN outperforms those on the other three dynamic PINs in terms of matching with known complexes, sensitivity, specificity, and f-measure. Furthermore, we evaluate the identified protein complexes by Gene Ontology (GO) function enrichment analysis. The validation shows that the identified protein complexes from ST-APIN are more biologically significant. This study provides a general paradigm for constructing the ST-APINs, which is essential for further understanding of molecular systems and the biomedical mechanism of complex diseases.

The methylation of DNA is an important mechanism to control biological processes. Recently, the Pacbio SMRT technology provides a new way to identify base methylation in the genome. MotifMaker is a tool developed by Pacbio for discovering DNA methylation motifs from methylated DNA sequences. However, MotifMaker is single-threaded and computational expensive for identifying methylation motifs from large genomes. Here, we present an efficient motif finding algorithm (MultiMotifMaker) by implementing multi threads of the MotifMaker. The MultiMotifMaker speeds up the motif search about 8-9 times on a 32 core computer comparing to MotifMaker. MultiMotifMaker makes it possible to identify methylation motifs from Pacbio reads for large genomes.