In this paper, we propose a novel framework, the Sampling-guided Heterogeneous Graph Neural Network (HT-GNN), to effectively tackle the challenge of missing data imputation in longitudinal studies. Unlike traditional methods, which often require extensive preprocessing to handle irregular or inconsistent missing data, our approach accommodates arbitrary missing data patterns while maintaining computational efficiency. HT-GNN models both observations and covariates as distinct node types, connecting observation nodes at successive time points through subject-specific longitudinal subnetworks, while covariate-observation interactions are represented by attributed edges within bipartite graphs. By leveraging subject-wise mini-batch sampling and a multi-layer temporal smoothing mechanism, HT-GNN efficiently scales to large datasets, while effectively learning node representations and imputing missing data. Extensive experiments on both synthetic and real-world datasets, including the Alzheimer's Disease Neuroimaging Initiative (DNI) dataset, demonstrate that HT-GNN significantly outperforms existing imputation methods, even with high missing data rates (e.g., 80%). The empirical results highlight HT-GNN's robust imputation capabilities and superior performance, particularly in the context of complex, large-scale longitudinal data.

Gene regulatory networks (GRNs) are critical blueprints for understanding gene regulation and the intricate interactions that drive biological processes. Recent advances have highlighted the potential of single-cell ATAC-seq (scATAC-seq) data in GRN inference, offering unprecedented insights into how chromatin accessibility plays an important part in gene regulation. However, existing methods often fall short in providing a quantitative and holistic depiction of regulatory relationships, particularly in capturing the strength, direction, and type of gene regulation simultaneously. In this paper, we present a novel approach that addresses these limitations by leveraging genetically perturbed scATAC-seq data to infer more comprehensive and accurate GRNs. Our method advances the field by integrating pre- and post-perturbation chromatin accessibility data, enabling the construction of GRNs that more accurately reflect the dynamic regulatory landscape. Through rigorous evaluation on seven real datasets, we demonstrate the method’s superior performance in reconstructing GRNs with enhanced precision and interpretability. This work significantly contributes to the field by providing a robust framework for GRN inference, with broad implications for understanding gene regulation in complex biological systems.

MOTIVATION: The prediction of cancer drug response is a challenging subject in modern personalized cancer therapy due to the uncertainty of drug efficacy and the heterogeneity of patients. It has been shown that the characteristics of the drug itself and the genomic characteristics of the patient can greatly influence the results of cancer drug response. Therefore, accurate, efficient, and comprehensive methods for drug feature extraction and genomics integration are crucial to improve the prediction accuracy. RESULTS: Accurate prediction of cancer drug response is vital for guiding the design of anticancer drugs. In this study, we propose an end-to-end deep learning model named DeepAEG which is based on a complete-graph update mode to predict IC50. Specifically, we integrate an edge update mechanism on the basis of a hybrid graph convolutional network to comprehensively learn the potential high-dimensional representation of topological structures in drugs, including atomic characteristics and chemical bond information. Additionally, we present a novel approach for enhancing simplified molecular input line entry specification data by employing sequence recombination to eliminate the defect of single sequence representation of drug molecules. Our extensive experiments show that DeepAEG outperforms other existing methods across multiple evaluation parameters in multiple test sets. Furthermore, we identify several potential anticancer agents, including bortezomib, which has proven to be an effective clinical treatment option. Our results highlight the potential value of DeepAEG in guiding the design of specific cancer treatment regimens.

Deciphering 3D genome conformation is important for understanding gene regulation and cellular function at a spatial level. The recent advances of single cell Hi-C technologies have enabled the profiling of the 3D architecture of DNA within individual cell, which allows us to study the cell-to-cell variability of 3D chromatin organization. Computational approaches are in urgent need to comprehensively analyze the sparse and heterogeneous single cell Hi-C data. Here, we proposed scDEC-Hi-C, a new framework for single cell Hi-C analysis with deep generative neural networks. scDEC-Hi-C outperforms existing methods in terms of single cell Hi-C data clustering and imputation. Moreover, the generative power of scDEC-Hi-C could help unveil the differences of chromatin architecture across cell types. We expect that scDEC-Hi-C could shed light on deepening our understanding of the complex mechanism underlying the formation of chromatin contacts.

MOTIVATION: Single-cell technologies play a crucial role in revolutionizing biological research over the past decade, which strengthens our understanding in cell differentiation, development and regulation from a single-cell level perspective. Single-cell RNA sequencing (scRNA-seq) is one of the most common single cell technologies, which enables probing transcriptional states in thousands of cells in one experiment. Identification of cell types from scRNA-seq measurements is a fundamental and crucial question to answer. Most previous studies directly take gene expression as input while ignoring the comprehensive gene-gene interactions. RESULTS: We propose scGraph, an automatic cell identification algorithm leveraging gene interaction relationships to enhance the performance of the cell-type identification. scGraph is based on a graph neural network to aggregate the information of interacting genes. In a series of experiments, we demonstrate that scGraph is accurate and outperforms eight comparison methods in the task of cell-type identification. Moreover, scGraph automatically learns the gene interaction relationships from biological data and the pathway enrichment analysis shows consistent findings with previous analysis, providing insights on the analysis of regulatory mechanism. AVAILABILITY AND IMPLEMENTATION: scGraph is freely available at https://github.com/QijinYin/scGraph and https://figshare.com/articles/software/scGraph/17157743. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Abstract Background Drug resistance is a critical obstacle in cancer therapy. Discovering cancer drug response is important to improve anti-cancer drug treatment and guide anti-cancer drug design. Abundant genomic and drug response resources of cancer cell lines provide unprecedented opportunities for such study. However, cancer cell lines cannot fully reflect heterogeneous tumor microenvironments. Transferring knowledge studied from in vitro cell lines to single-cell and clinical data will be a promising direction to better understand drug resistance. Most current studies include single nucleotide variants (SNV) as features and focus on improving predictive ability of cancer drug response on cell lines. However, obtaining accurate SNVs from clinical tumor samples and single-cell data is not reliable. This makes it difficult to generalize such SNV-based models to clinical tumor data or single-cell level studies in the future. Results We present a new method, DualGCN, a unified Dual Graph Convolutional Network model to predict cancer drug response. DualGCN encodes both chemical structures of drugs and omics data of biological samples using graph convolutional networks. Then the two embeddings are fed into a multilayer perceptron to predict drug response. DualGCN incorporates prior knowledge on cancer-related genes and protein–protein interactions, and outperforms most state-of-the-art methods while avoiding using large-scale SNV data. Conclusions The proposed method outperforms most state-of-the-art methods in predicting cancer drug response without the use of large-scale SNV data. These favorable results indicate its potential to be extended to clinical and single-cell tumor samples and advancements in precision medicine.

Federated learning (FL) is a privacy-preserving paradigm for multi-institutional collaborations, where the aggregation is an essential procedure after training on the local datasets. Conventional aggregation algorithms often apply a weighted averaging of the updates generated from distributed machines to update the global model. However, while the data distributions are non-IID, the large discrepancy between the local updates might lead to a poor averaged result and a lower convergence speed, i.e., more iterations required to achieve a certain performance. To solve this problem, this article proposes a novel method named AggEnhance for enhancing the aggregation, where we synthesize a group of reliable samples from the local models and tune the aggregated result on them. These samples, named class interior points (CIPs) in this work, bound the relevant decision boundaries that ensure the performance of aggregated result. To the best of our knowledge, this is the first work to explicitly design an enhancing method for the aggregation in prevailing FL pipelines. A series of experiments on real data demonstrate that our method has noticeable improvements of the convergence in non-IID scenarios. In particular, our approach reduces the iterations by 31.87% on average for the CIFAR10 dataset and 43.90% for the PASCAL VOC dataset. Since our method does not modify other procedures of FL pipelines, it is easy to apply to most existing FL frameworks. Furthermore, it does not require additional data transmitted from the local clients to the global server, thus holding the same security level as the original FL algorithms.

Training multiple deep neural networks (DNNs) and averaging their outputs is a simple way to improve the predictive performance. Nevertheless, the multiplied training cost prevents this ensemble method to be practical and efficient. Several recent works attempt to save and ensemble the checkpoints of DNNs, which only requires the same computational cost as training a single network. However, these methods suffer from either marginal accuracy improvements due to the low diversity of checkpoints or high risk of divergence due to the cyclical learning rates they adopted. In this paper, we propose a novel method to ensemble the checkpoints, where a boosting scheme is utilized to accelerate model convergence and maximize the checkpoint diversity. We theoretically prove that it converges by reducing exponential loss. The empirical evaluation also indicates our proposed ensemble outperforms single model and existing ensembles in terms of accuracy and efficiency. With the same training budget, our method achieves 4.16% lower error on Cifar-100 and 6.96% on Tiny-ImageNet with ResNet-110 architecture. Moreover, the adaptive sample weights in our method make it an effective solution to address the imbalanced class distribution. In the experiments, it yields up to 5.02% higher accuracy over single EfficientNet-B0 on the imbalanced datasets.

A major challenge in the pharmaceutical industry is to design novel molecules with specific desired properties, especially when the property evaluation is costly. Here, we propose MNCE-RL, a graph convolutional policy network for molecular optimization with molecular neighborhood-controlled embedding grammars through reinforcement learning. We extend the original neighborhood-controlled embedding grammars to make them applicable to molecular graph generation and design an efficient algorithm to infer grammatical production rules from given molecules. The use of grammars guarantees the validity of the generated molecular structures. By transforming molecular graphs to parse trees with the inferred grammars, the molecular structure generation task is modeled as a Markov decision process where a policy gradient strategy is utilized. In a series of experiments, we demonstrate that our approach achieves state-of-the-art performance in a diverse range of molecular optimization tasks and exhibits significant superiority in optimizing molecular properties with a limited number of property evaluations.

MOTIVATION: Accurate prediction of cancer drug response (CDR) is challenging due to the uncertainty of drug efficacy and heterogeneity of cancer patients. Strong evidences have implicated the high dependence of CDR on tumor genomic and transcriptomic profiles of individual patients. Precise identification of CDR is crucial in both guiding anti-cancer drug design and understanding cancer biology. RESULTS: In this study, we present DeepCDR which integrates multi-omics profiles of cancer cells and explores intrinsic chemical structures of drugs for predicting CDR. Specifically, DeepCDR is a hybrid graph convolutional network consisting of a uniform graph convolutional network and multiple subnetworks. Unlike prior studies modeling hand-crafted features of drugs, DeepCDR automatically learns the latent representation of topological structures among atoms and bonds of drugs. Extensive experiments showed that DeepCDR outperformed state-of-the-art methods in both classification and regression settings under various data settings. We also evaluated the contribution of different types of omics profiles for assessing drug response. Furthermore, we provided an exploratory strategy for identifying potential cancer-associated genes concerning specific cancer types. Our results highlighted the predictive power of DeepCDR and its potential translational value in guiding disease-specific drug design. AVAILABILITY AND IMPLEMENTATION: DeepCDR is freely available at https://github.com/kimmo1019/DeepCDR. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: Advances in high-throughput genotyping and sequencing technologies during recent years have revealed essential roles of non-coding regions in gene regulation. Genome-wide association studies (GWAS) suggested that a large proportion of risk variants are located in non-coding regions and remain unexplained by current expression quantitative trait loci catalogs. Interpreting the causal effects of these genetic modifications is crucial but difficult owing to our limited knowledge of how regulatory elements function. Although several computational methods have been designed to prioritize regulatory variants that substantially impact human phenotypes, few of them achieve consistently high performance even when large-scale multi-omic data are integrated. RESULTS: We propose a novel multi-task framework based on Bayesian deep neural networks, MtBNN, to quantify the deleterious impact of single nucleotide polymorphisms in non-coding genomic regions. With the high-efficiency provided by the multi-task Bayesian framework to integrate information from different sources, MtBNN is capable of extracting features from genomic sequences of large-scale chromatin-profiling data, such as chromatin accessibility and transcript factor binding affinities, and calculating the distribution of the probability that a non-coding variant disrupts regulatory activities. A series of comprehensive experiments show that MtBNN quantifies the functional impact of cis-regulatory variations with high accuracy, including expression quantitative trait locus, DNase I sensitivity quantitative trait locus and functional genetic variants located within ATAC-peaks that affect the accessibility of the corresponding peak and achieves significantly better performance than the existing methods. Moreover, MtBNN has applications in the discovery of potentially causal disease-associated single-nucleotide polymorphisms (SNPs), thus helping fine-map the GWAS SNPs. AVAILABILITY AND IMPLEMENTATION: Code can be downloaded from https://github.com/Zoesgithub/MtBNN. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: Hi-C is a genome-wide technology for investigating 3D chromatin conformation by measuring physical contacts between pairs of genomic regions. The resolution of Hi-C data directly impacts the effectiveness and accuracy of downstream analysis such as identifying topologically associating domains (TADs) and meaningful chromatin loops. High resolution Hi-C data are valuable resources which implicate the relationship between 3D genome conformation and function, especially linking distal regulatory elements to their target genes. However, high resolution Hi-C data across various tissues and cell types are not always available due to the high sequencing cost. It is therefore indispensable to develop computational approaches for enhancing the resolution of Hi-C data. RESULTS: We proposed hicGAN, an open-sourced framework, for inferring high resolution Hi-C data from low resolution Hi-C data with generative adversarial networks (GANs). To the best of our knowledge, this is the first study to apply GANs to 3D genome analysis. We demonstrate that hicGAN effectively enhances the resolution of low resolution Hi-C data by generating matrices that are highly consistent with the original high resolution Hi-C matrices. A typical scenario of usage for our approach is to enhance low resolution Hi-C data in new cell types, especially where the high resolution Hi-C data are not available. Our study not only presents a novel approach for enhancing Hi-C data resolution, but also provides fascinating insights into disclosing complex mechanism underlying the formation of chromatin contacts. AVAILABILITY AND IMPLEMENTATION: We release hicGAN as an open-sourced software at https://github.com/kimmo1019/hicGAN. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Motivation: A majority of known genetic variants associated with human-inherited diseases lie in non-coding regions that lack adequate interpretation, making it indispensable to systematically discover functional sites at the whole genome level and precisely decipher their implications in a comprehensive manner. Although computational approaches have been complementing high-throughput biological experiments towards the annotation of the human genome, it still remains a big challenge to accurately annotate regulatory elements in the context of a specific cell type via automatic learning of the DNA sequence code from large-scale sequencing data. Indeed, the development of an accurate and interpretable model to learn the DNA sequence signature and further enable the identification of causative genetic variants has become essential in both genomic and genetic studies. Results: We proposed Deopen, a hybrid framework mainly based on a deep convolutional neural network, to automatically learn the regulatory code of DNA sequences and predict chromatin accessibility. In a series of comparison with existing methods, we show the superior performance of our model in not only the classification of accessible regions against background sequences sampled at random, but also the regression of DNase-seq signals. Besides, we further visualize the convolutional kernels and show the match of identified sequence signatures and known motifs. We finally demonstrate the sensitivity of our model in finding causative noncoding variants in the analysis of a breast cancer dataset. We expect to see wide applications of Deopen with either public or in-house chromatin accessibility data in the annotation of the human genome and the identification of non-coding variants associated with diseases. Availability and implementation: Deopen is freely available at https://github.com/kimmo1019/Deopen. Contact: [email protected]. Supplementary information: Supplementary data are available at Bioinformatics online.