Retrieving targeted pathways in biological knowledge bases, particularly when incorporating wet-lab experimental data, remains a challenging task and often requires downstream analyses and specialized expertise. In this paper, we frame this challenge as a solvable graph learning and explaining task and propose a novel subgraph inference framework, ExPath, that explicitly integrates experimental data to classify various graphs (bio-networks) in biological databases. The links (representing pathways) that contribute more to classification can be considered as targeted pathways. Our framework can seamlessly integrate biological foundation models to encode the experimental molecular data. We propose ML-oriented biological evaluations and a new metric. The experiments involving 301 bio-networks evaluations demonstrate that pathways inferred by ExPath are biologically meaningful, achieving up to 4.5× higher Fidelity+ (necessity) and 14× lower Fidelity- (sufficiency) than explainer baselines, while preserving signaling chains up to 4× longer.

Deep learning models have recently shown great success in classifying epileptic patients using EEG recordings. Unfortunately, classification-based methods lack a sound mechanism to detect the onset of seizure events. In this work, we propose a two-stage framework, SODor, that explicitly models seizure onset through a novel task formulation of subsequence clustering. Given an EEG sequence, the framework first learns a set of second-level embeddings with label supervision. It then employs model-based clustering to explicitly capture long-term temporal dependencies in EEG sequences and identify meaningful subsequences. Epochs within a subsequence share a common cluster assignment (normal or seizure), with cluster or state transitions representing successful onset detections. Extensive experiments on three datasets demonstrate that our method can correct misclassifications, achieving 5%-11% classification improvements over other baselines and accurately detecting seizure onsets.

Retrieving gene functional networks from knowledge databases presents a challenge due to the mismatch between disease networks and subtype-specific variations. Current solutions, including statistical and deep learning methods, often fail to effectively integrate gene interaction knowledge from databases or explicitly learn subtype-specific interactions. To address this mismatch, we propose GeSubNet, which learns a unified representation capable of predicting gene interactions while distinguishing between different disease subtypes. Graphs generated by such representations can be considered subtype-specific networks. GeSubNet is a multi-step representation learning framework with three modules: First, a deep generative model learns distinct disease subtypes from patient gene expression profiles. Second, a graph neural network captures representations of prior gene networks from knowledge databases, ensuring accurate physical gene interactions. Finally, we integrate these two representations using an inference loss that leverages graph generation capabilities, conditioned on the patient separation loss, to refine subtype-specific information in the learned representation. GeSubNet consistently outperforms traditional methods, with average improvements of 30.6%, 21.0%, 20.1%, and 56.6% across four graph evaluation metrics, averaged over four cancer datasets. Particularly, we conduct a biological simulation experiment to assess how the behavior of selected genes from over 11,000 candidates affects subtypes or patient distributions. The results show that the generated network has the potential to identify subtype-specific genes with an 83% likelihood of impacting patient distribution shifts.

Dynamic GNNs, which integrate temporal and spatial features in Electroencephalography (EEG) data, have shown great potential in automating seizure detection. However, fully capturing the underlying dynamics necessary to represent brain states, such as seizure and non-seizure, remains a non-trivial task and presents two fundamental challenges. First, most existing dynamic GNN methods are built on temporally fixed static graphs, which fail to reflect the evolving nature of brain connectivity during seizure progression. Second, current efforts to jointly model temporal signals and graph structures and, more importantly, their interactions remain nascent, often resulting in inconsistent performance. To address these challenges, we present the first theoretical analysis of these two problems, demonstrating the effectiveness and necessity of explicit dynamic modeling and time-then-graph dynamic GNN method. Building on these insights, we propose EvoBrain, a novel seizure detection model that integrates a two-stream Mamba architecture with a GCN enhanced by Laplacian Positional Encoding, following neurological insights. Moreover, EvoBrain incorporates explicitly dynamic graph structures, allowing both nodes and edges to evolve over time. Our contributions include (a) a theoretical analysis proving the expressivity advantage of explicit dynamic modeling and time-then-graph over other approaches, (b) a novel and efficient model that significantly improves AUROC by 23\% and F1 score by 30\%, compared with the dynamic GNN baseline, and (c) broad evaluation of our method on the challenging early seizure prediction task.

While end-to-end multi-channel electroencephalography (EEG) learning approaches have shown significant promise, their applicability is often constrained in neurological diagnostics, such as intracranial EEG resources. When provided with a single-channel EEG, how can we learn representations that are robust to multi-channels and scalable across varied tasks, such as seizure prediction? In this paper, we present SplitSEE, a structurally splittable framework designed for effective temporal-frequency representation learning in single-channel EEG. The key concept of SplitSEE is a self-supervised framework incorporating a deep clustering task. Given an EEG, we argue that the time and frequency domains are two distinct perspectives, and hence, learned representations should share the same cluster assignment. To this end, we first propose two domain-specific modules that independently learn domain-specific representation and address the temporal-frequency tradeoff issue in conventional spectrogram-based methods. Then, we introduce a novel clustering loss to measure the information similarity. This encourages representations from both domains to coherently describe the same input by assigning them a consistent cluster. SplitSEE leverages a pretraining-to-fine-tuning framework within a splittable architecture and has following properties: (a) Effectiveness: it learns representations solely from single-channel EEG but has even outperformed multi-channel baselines. (b) Robustness: it shows the capacity to adapt across different channels with low performance variance. Superior performance is also achieved with our collected clinical dataset. (c) Scalability: With just one fine-tuning epoch, SplitSEE achieves high and stable performance using partial model layers.

The Transformer model has shown leading performance in time series forecasting. Nevertheless, in some complex scenarios, it tends to learn low-frequency features in the data and overlook high-frequency features, showing a frequency bias. This bias prevents the model from accurately capturing important high-frequency data features. In this paper, we undertake empirical analyses to understand this bias and discover that frequency bias results from the model disproportionately focusing on frequency features with higher energy. Based on our analysis, we formulate this bias and propose Fredformer, a Transformer-based framework designed to mitigate frequency bias by learning features equally across different frequency bands. This approach prevents the model from overlooking lower amplitude features important for accurate forecasting. Extensive experiments show the effectiveness of our proposed approach, which can outperform other baselines in different real-world time-series datasets. Furthermore, we introduce a lightweight variant of the Fredformer with an attention matrix approximation, which achieves comparable performance but with much fewer parameters and lower computation costs. The code is available at: https://github.com/chenzRG/Fredformer

Deep neural networks have demonstrated impressive results in medical image analysis, but designing suitable architectures for each specific task is expertise-dependent and time-consuming. Neural architecture search (NAS) offers an effective means of discovering architectures. It has been highly successful in numerous applications, particularly in natural image classification. Yet, medical images possess unique characteristics, such as small regions and a wide variety of lesion sizes, that differentiate them from natural images. Furthermore, most current NAS methods struggle with high computational costs, especially when dealing with high-resolution image datasets. In this paper, we present a novel evolutionary neural architecture search method called Multi-Scale Training-Free Neural Architecture Search to address these challenges. Specifically, to accommodate the broad range of lesion region sizes in disease diagnosis, we develop a new reduction cell search space that enables the search algorithm to explicitly identify the optimal scale combination for multi-scale feature extraction. To overcome the issue of high computational costs, we utilize training-free indicators as performance measures for candidate architectures, which allows us to search for the optimal architecture more efficiently. More specifically, by considering the capability and simplicity of various networks, we formulate a multi-objective optimization problem that involves two training-free indicators and model complexity for candidate architectures. Extensive experiments on a large medical image benchmark and a publicly available breast cancer detection dataset are conducted. The empirical results demonstrate that our MSTF-NAS outperforms both human-designed architectures and current state-of-the-art NAS algorithms on both datasets, indicating the effectiveness of our proposed method.

Comprehensive analysis of multi-omics data has now garnered significant attention. However, due to the diversity of multi-omics data, integrating multi-omics information presents a formidable challenge for researchers. Moreover, the high dimensionality and sparsity characteristics in omics data further complicate multi-omics data analysis. To address these challenges and obtain high-quality representations suitable for downstream tasks, we propose a self-supervised clustering learning framework called the Multi-view Feature Hierarchical Contrastive Clustering model (MFHCC) to extract multi-level features. Firstly, the proposed model considers multi-omics as multi-modality and employs an autoencoder for each modality to integrate diverse omics information simultaneously. Secondly, it utilizes a multilevel feature extraction framework with contrastive learning methods to mitigate the impact of redundant information and null values on representation quality while capturing semantic information embedded in the data. Additionally, the model incorporates a deep clustering module to guide the representation toward downstream tasks while integrating high-level features for guidance. Through extensive experiments conducted on pan-cancer datasets, we validate the effectiveness of MFHCC. For instance, the model achieves an accuracy exceeding 76% by omics types, thus confirming its superior performance.

Many policy optimization approaches in reinforcement learning incorporate a Kullback-Leilbler (KL) divergence to the previous policy, to prevent the policy from changing too quickly. This idea was initially proposed in a seminal paper on Conservative Policy Iteration, with approximations given by algorithms like TRPO and Munchausen Value Iteration (MVI). We continue this line of work by investigating a generalized KL divergence---called the Tsallis KL divergence. Tsallis KL defined by the $q$-logarithm is a strict generalization, as $q = 1$ corresponds to the standard KL divergence; $q > 1$ provides a range of new options. We characterize the types of policies learned under the Tsallis KL, and motivate when $q >1$ could be beneficial. To obtain a practical algorithm that incorporates Tsallis KL regularization, we extend MVI, which is one of the simplest approaches to incorporate KL regularization. We show that this generalized MVI($q$) obtains significant improvements over the standard MVI($q = 1$) across 35 Atari games.

Electroencephalography (EEG) is a major tool for studying neurophysiological processes. Investigating reliable representations from highly noisy measurements is a pending challenge, however, the medically treasured and insufficient labeled data have driven this process away from a supervised learning manner. Recent works have turned their attention to self-supervised learning (SSL), putting the contrastive strategy on capturing the spatio-temporal characteristics of the neuronal events of interest. We argue that the temporal-spatial view is not the best choice for the SSL contrastive objective because there is a missing piece of the EEG representation that is usually ignored: dynamic fluctuations in brain neurons and the statistical learning of analog/artificial neural networks cannot handle the dynamic characteristics well. This paper proposes a novel two-view contrastive learning framework to refine EEG features from local-global and past-future views. An array of spiking neural networks is embedded to project spatio-temporal features onto the spike sequences to represent the dynamic fluctuation information of EEG. Experimenting with sleep stage classification and prediction of lethal epileptic seizures, we verify the proposal competes favorably against the state-of-the-art methods and offers high-quality features, that is, supervised learning on top of them observes a significant improvement in classification after only one training iteration.

Identifying a specific cancer subtype from a variety of candidates is vital for precise and effective treatment. However, cancer subtyping is highly non-trivial as a result of cancer heterogeneity. While significant efforts have been put into understanding the mechanism of cancer subtypes via studying the omics data, existing methods run the risk of presenting biased analyses resulted from overfitting the high-dimensional and scarce omics data. In this paper, we propose a novel generative model that directly models the cancer data distribution by which downstream tasks can circumvent the curse of overfitting and achieve better performance. Unlike conventional generative modeling schemes, the proposed method underlines hierarchical categorical latent spaces to extract global features and local details respectively from transcriptomics and genomics profiles, which is the first to be considered in the cancer subtyping literature. By extensive experiments we verify that the proposed architecture achieves more clearly separated subtypes, as well as medically significant insights into real subtyping.

An end-to-end platform assembling multiple tiers is built for precisely cognizing brain activities. Being fed massive electroencephalogram (EEG) data, the time-frequency spectrograms are conventionally projected into the episode-wise feature matrices (seen as tier-1). A spiking neural network (SNN) based tier is designed to distill the principle information in terms of spike-streams from the rare features, which maintains the temporal implication in the nature of EEGs. The proposed tier-3 transposes time- and space-domain of spike patterns from the SNN; and feeds the transposed pattern-matrices into an artificial neural network (ANN, Transformer specifically) known as tier-4, where a special spanning topology is proposed to match the two-dimensional input form. In this manner, cognition such as classification is conducted with high accuracy. For proof-of-concept, the sleep stage scoring problem is demonstrated by introducing multiple EEG datasets with the largest comprising 42,560 hours recorded from 5,793 subjects. From experiment results, our platform achieves the general cognition overall accuracy of 87% by leveraging sole EEG, which is 2% superior to the state-of-the-art. Moreover, our developed multi-tier methodology offers visible and graphical interpretations of the temporal characteristics of EEG by identifying the critical episodes, which is demanded in neurodynamics but hardly appears in conventional cognition scenarios.

Sleep screening is not only a major tool in the assessment of pathophysiology, but also a bridge between the central neuronal systems and behaviour/cognition. Automatic sleep staging is an alternative for the time-consuming gold standard manual scoring procedure. Most of the existing works designed such procedure by using deep neural networks without considering the medical criterion of the sleep staging task. We argue that capturing the stage-specific features which meet the criterion is of significant importance for the automatic sleep staging alternative. In this work we propose an end-to-end sleep staging simulator based on mixed neural networks, i.e., CNN, LSTM, and Transformer. The framework consists of two subnetworks: stage dependent feature mapping network which is constructed by the idea of physiological sleep nature, and an attention-based parallel staging network. Moreover, we adopt a mixed precision training strategy to quantize the model for exploring feasible usage in the clinical settings. Through an experiment with a large EEG database (Sleep Heart Health Study), the proposed method has a competitive stage scoring performance, especially in stages Wake, N2, and N3, with higher precision of 0.92, 0.85, and 0.86, respectively. Our study proves that the quantized model has potential capability for further application in the clinical staging task.

The gut microbiota is crucial for human physiology and susceptibility to diseases. Knowing the AMR phenotype canfacilitate the understanding of the impact of antibiotics administration on the gut microbiota. Nowadays, whole-genome sequencing for antibiotic susceptibility testing (WGS-AST) is widely used in clinical microbiology to predict the AMR phenotype. To release the limitations of the genomic information and improve the WGS-AST prediction, we propose an integrated multi-omics approach, employing a deep generative neural network (VAE: variational auto-encoder). We evaluate the proposed approach by two machine learning techniques (i.e., K-means for clustering and Random Forest for classification). Our evaluation results show that the integrated multi-omics approach achieves relatively better performance than the conventional WGS-AST. Moreover, the integrated multi-omics approach is able to visually reveal AMR phenotype of the gutmicrobiota via antibacterial spectrum. Our work provides evidence that multi-omics information is useful to enhance the WGS-AST prediction.

Recently, deep learning-based methods have been successfully proposed for electrophysiology signal-based sleep staging with promising results. Most existing methods use convolutional layers and recurrent-based architectures to implement a model structure from feature extraction to sequence signal classification. In this study, we propose a method of segmenting electroencephalogram (EEG) and electrooculogram (EOG) data according to frequency bands and construct a Transformer based automatic sleep classification model on top of it. The results show that the classifications of the stage Wake, N3, and REM outperform the state-of-art works, with the Fl-scores of 0.92, 0.85 and 0.91. Our work is the first attempt to explore the feasibility of a truth-involved Transformer-based model with a large-scale sleep database.

Due to the inherent variability such as inter-users anatomical variability and the inter-systems differences, the design of new EEG-based index and a reliable model for sleep stages classification is still the main topic in sleep science. The unsupervised deep learning framework-variational autoencoder (VAE) which can capture the major characteristics of the input by imposing a Gaussian prior distribution on the latent features is suitable in EEG characterization and reconstruction. Although vanilla VAE and convolutional autoencoder (CAE) have been tried, it has yet been discussed that whether a deep structure or a multi-scale structure is more appropriate. In this paper, we constructed a shallow iVAE model, which will capture the multi-scale features of the spectrogram of EEG by replacing the main structure in encoder and decoder with the inception-like structure. By comparing with the vanilla VAE and the CAE, a more accurate reconstruction and a better classification using the latent features of the iVAE can be confirmed.