High-quality genome chromosome-scale sequences provide an important basis for genomics downstream analysis, especially the construction of haplotype-resolved and complete genomes, which plays a key role in genome annotation, mutation detection, evolutionary analysis, gene function research, comparative genomics and other aspects. However, genome-wide short-read sequencing is difficult to produce a complete genome in the face of a complex genome with high duplication and multiple heterozygosity. The emergence of long-read sequencing technology has greatly improved the integrity of complex genome assembly. We review a variety of computational methods for complex genome assembly and describe in detail the theories, innovations and shortcomings of collapsed, semi-collapsed and uncollapsed assemblers based on long reads. Among the three methods, uncollapsed assembly is the most correct and complete way to represent genomes. In addition, genome assembly is closely related to haplotype reconstruction, that is uncollapsed assembly realizes haplotype reconstruction, and haplotype reconstruction promotes uncollapsed assembly. We hope that gapless, telomere-to-telomere and accurate assembly of complex genomes can be truly routinely achieved using only a simple process or a single tool in the future.

Agglomerative hierarchical clustering has become a common tool for the analysis and visualization of data, thus being present in a large amount of scientific research and predating all areas of bioinformatics and computational biology. In this work, we focus on a critical problem, the nonuniqueness of the clustering when there are tied distances, for which several solutions exist but are not implemented in most hierarchical clustering packages. We analyze the magnitude of this problem in one particular setting: the clustering of microsatellite markers using the Unweighted Pair-Group Method with Arithmetic Mean. To do so, we have calculated the fraction of publications at the Scopus database in which more than one hierarchical clustering is possible, showing that about 46% of the articles are affected. Additionally, to show the problem from a practical point of view, we selected two opposite examples of articles that have multiple solutions: one with two possible dendrograms, and the other with more than 2.5 million different possible hierarchical clusterings.

Many computational methods are devoted to rapidly generating pseudo-natural products to expand the open-ended border of chemical spaces for natural products. However, the accessibility and chemical interpretation were often ignored or underestimated in conventional library/fragment-based or rule-based strategies, thus hampering experimental synthesis. Herein, a bio-inspired strategy (named TeroGen) is developed to mimic the two key biosynthetic stages (cyclization and decoration) of terpenoid natural products, by utilizing physically based simulations and deep learning models, respectively. The precision and efficiency are validated for different categories of terpenoids, and in practice, more than 30 000 sesterterpenoids (10 times as many as the known sesterterpenoids) are predicted to be linked in a reaction network, and their synthetic accessibility and chemical interpretation are estimated by thermodynamics and kinetics. Since it could not only greatly expand the chemical space of terpenoids but also numerate plausible biosynthetic routes, TeroGen is promising for accelerating heterologous biosynthesis, bio-mimic and chemical synthesis of complicated terpenoids and derivatives.

Antibiotic combination is a promising strategy to extend the lifetime of antibiotics and thereby combat antimicrobial resistance. However, screening for new antibiotic combinations is both time-consuming and labor-intensive. In recent years, an increasing number of researchers have used computational models to predict effective antibiotic combinations. In this review, we summarized existing computational models for antibiotic combinations and discussed the limitations and challenges of these models in detail. In addition, we also collected and summarized available data resources and tools for antibiotic combinations. This study aims to help computational biologists design more accurate and interpretable computational models.

Intraclonal diversification (ID) within the immunoglobulin (IG) genes expressed by B cell clones arises due to ongoing somatic hypermutation (SHM) in a context of continuous interactions with antigen(s). Defining the nature and order of appearance of SHMs in the IG genes can assist in improved understanding of the ID process, shedding light into the ontogeny and evolution of B cell clones in health and disease. Such endeavor is empowered thanks to the introduction of high-throughput sequencing in the study of IG gene repertoires. However, few existing tools allow the identification, quantification and characterization of SHMs related to ID, all of which have limitations in their analysis, highlighting the need for developing a purpose-built tool for the comprehensive analysis of the ID process. In this work, we present the immunoglobulin intraclonal diversification analysis (IgIDivA) tool, a novel methodology for the in-depth qualitative and quantitative analysis of the ID process from high-throughput sequencing data. IgIDivA identifies and characterizes SHMs that occur within the variable domain of the rearranged IG genes and studies in detail the connections between identified SHMs, establishing mutational pathways. Moreover, it combines established and new graph-based metrics for the objective determination of ID level, combined with statistical analysis for the comparison of ID level features for different groups of samples. Of importance, IgIDivA also provides detailed visualizations of ID through the generation of purpose-built graph networks. Beyond the method design, IgIDivA has been also implemented as an R Shiny web application. IgIDivA is freely available at https://bio.tools/igidiva.

In a drug formulation (DFM), the major components by mass are not Active Pharmaceutical Ingredient (API) but rather Drug Inactive Ingredients (DIGs). DIGs can reach much higher concentrations than that achieved by API, which raises great concerns about their clinical toxicities. Therefore, the biological activities of DIG on physiologically relevant target are widely demanded by both clinical investigation and pharmaceutical industry. However, such activity data are not available in any existing pharmaceutical knowledge base, and their potentials in predicting the DIG-target interaction have not been evaluated yet. In this study, the comprehensive assessment and analysis on the biological activities of DIGs were therefore conducted. First, the largest number of DIGs and DFMs were systematically curated and confirmed based on all drugs approved by US Food and Drug Administration. Second, comprehensive activities for both DIGs and DFMs were provided for the first time to pharmaceutical community. Third, the biological targets of each DIG and formulation were fully referenced to available databases that described their pharmaceutical/biological characteristics. Finally, a variety of popular artificial intelligence techniques were used to assess the predictive potential of DIGs' activity data, which was the first evaluation on the possibility to predict DIG's activity. As the activities of DIGs are critical for current pharmaceutical studies, this work is expected to have significant implications for the future practice of drug discovery and precision medicine.

Proteogenomics refers to the integrated analysis of the genome and proteome that leverages mass-spectrometry (MS)-based proteomics data to improve genome annotations, understand gene expression control through proteoforms and find sequence variants to develop novel insights for disease classification and therapeutic strategies. However, proteogenomic studies often suffer from reduced sensitivity and specificity due to inflated database size. To control the error rates, proteogenomics depends on the target-decoy search strategy, the de-facto method for false discovery rate (FDR) estimation in proteomics. The proteogenomic databases constructed from three- or six-frame nucleotide database translation not only increase the search space and compute-time but also violate the equivalence of target and decoy databases. These searches result in poorer separation between target and decoy scores, leading to stringent FDR thresholds. Understanding these factors and applying modified strategies such as two-pass database search or peptide-class-specific FDR can result in a better interpretation of MS data without introducing additional statistical biases. Based on these considerations, a user can interpret the proteogenomics results appropriately and control false positives and negatives in a more informed manner. In this review, first, we briefly discuss the proteogenomic workflows and limitations in database construction, followed by various considerations that can influence potential novel discoveries in a proteogenomic study. We conclude with suggestions to counter these challenges for better proteogenomic data interpretation.

Several new viral infections have emerged in the human population and establishing as global pandemics. With advancements in translation research, the scientific community has developed potential therapeutics to eradicate or control certain viral infections, such as smallpox and polio, responsible for billions of disabilities and deaths in the past. Unfortunately, some viral infections, such as dengue virus (DENV) and human immunodeficiency virus-1 (HIV-1), are still prevailing due to a lack of specific therapeutics, while new pathogenic viral strains or variants are emerging because of high genetic recombination or cross-species transmission. Consequently, to combat the emerging viral infections, bioinformatics-based potential strategies have been developed for viral characterization and developing new effective therapeutics for their eradication or management. This review attempts to provide a single platform for the available wide range of bioinformatics-based approaches, including bioinformatics methods for the identification and management of emerging or evolved viral strains, genome analysis concerning the pathogenicity and epidemiological analysis, computational methods for designing the viral therapeutics, and consolidated information in the form of databases against the known pathogenic viruses. This enriched review of the generally applicable viral informatics approaches aims to provide an overview of available resources capable of carrying out the desired task and may be utilized to expand additional strategies to improve the quality of translation viral informatics research.

Updated and expert-quality knowledge bases are fundamental to biomedical research. A knowledge base established with human participation and subject to multiple inspections is needed to support clinical decision making, especially in the growing field of precision oncology. The number of original publications in this field has risen dramatically with the advances in technology and the evolution of in-depth research. Consequently, the issue of how to gather and mine these articles accurately and efficiently now requires close consideration. In this study, we present OncoPubMiner (https://oncopubminer.chosenmedinfo.com), a free and powerful system that combines text mining, data structure customisation, publication search with online reading and project-centred and team-based data collection to form a one-stop 'keyword in-knowledge out' oncology publication mining platform. The platform was constructed by integrating all open-access abstracts from PubMed and full-text articles from PubMed Central, and it is updated daily. OncoPubMiner makes obtaining precision oncology knowledge from scientific articles straightforward and will assist researchers in efficiently developing structured knowledge base systems and bring us closer to achieving precision oncology goals.

Automated relation extraction (RE) from biomedical literature is critical for many downstream text mining applications in both research and real-world settings. However, most existing benchmarking datasets for biomedical RE only focus on relations of a single type (e.g. protein-protein interactions) at the sentence level, greatly limiting the development of RE systems in biomedicine. In this work, we first review commonly used named entity recognition (NER) and RE datasets. Then, we present a first-of-its-kind biomedical relation extraction dataset (BioRED) with multiple entity types (e.g. gene/protein, disease, chemical) and relation pairs (e.g. gene-disease; chemical-chemical) at the document level, on a set of 600 PubMed abstracts. Furthermore, we label each relation as describing either a novel finding or previously known background knowledge, enabling automated algorithms to differentiate between novel and background information. We assess the utility of BioRED by benchmarking several existing state-of-the-art methods, including Bidirectional Encoder Representations from Transformers (BERT)-based models, on the NER and RE tasks. Our results show that while existing approaches can reach high performance on the NER task (F-score of 89.3%), there is much room for improvement for the RE task, especially when extracting novel relations (F-score of 47.7%). Our experiments also demonstrate that such a rich dataset can successfully facilitate the development of more accurate, efficient and robust RE systems for biomedicine. Availability: The BioRED dataset and annotation guidelines are freely available at https://ftp.ncbi.nlm.nih.gov/pub/lu/BioRED/.

Identifying genomic regions influenced by natural selection provides fundamental insights into the genetic basis of local adaptation. However, it remains challenging to detect loci under complex spatially varying selection. We propose a deep learning-based framework, DeepGenomeScan, which can detect signatures of spatially varying selection. We demonstrate that DeepGenomeScan outperformed principal component analysis- and redundancy analysis-based genome scans in identifying loci underlying quantitative traits subject to complex spatial patterns of selection. Noticeably, DeepGenomeScan increases statistical power by up to 47.25% under nonlinear environmental selection patterns. We applied DeepGenomeScan to a European human genetic dataset and identified some well-known genes under selection and a substantial number of clinically important genes that were not identified by SPA, iHS, Fst and Bayenv when applied to the same dataset.

Integration of accumulative large-scale single-cell transcriptomes requires scalable batch-correction approaches. Here we propose Fugue, a simple and efficient batch-correction method that is scalable for integrating super large-scale single-cell transcriptomes from diverse sources. The core idea of the method is to encode batch information as trainable parameters and add it to single-cell expression profile; subsequently, a contrastive learning approach is used to learn feature representation of the additive expression profile. We demonstrate the scalability of Fugue by integrating all single cells obtained from the Human Cell Atlas. We benchmark Fugue against current state-of-the-art methods and show that Fugue consistently achieves improved performance in terms of data alignment and clustering preservation. Our study will facilitate the integration of single-cell transcriptomes at increasingly large scale.

Protein side chains are vitally important to many biological processes such as protein-protein interaction. In this study, we evaluate the performance of our previous released side-chain modeling method OPUS-Mut, together with some other methods, on three oligomer datasets, CASP14 (11), CAMEO-Homo (65) and CAMEO-Hetero (21). The results show that OPUS-Mut outperforms other methods measured by all residues or by the interfacial residues. We also demonstrate our method on evaluating protein-protein docking pose on a dataset Oligomer-Dock (75) created using the top 10 predictions from ZDOCK 3.0.2. Our scoring function correctly identifies the native pose as the top-1 in 45 out of 75 targets. Different from traditional scoring functions, our method is based on the overall side-chain packing favorableness in accordance with the local packing environment. It emphasizes the significance of side chains and provides a new and effective scoring term for studying protein-protein interaction.

Cells and tissues respond to perturbations in multiple ways that can be sensitively reflected in the alterations of gene expression. Current approaches to finding and quantifying the effects of perturbations on cell-level responses over time disregard the temporal consistency of identifiable gene programs. To leverage the occurrence of these patterns for perturbation analyses, we developed CellDrift (https://github.com/KANG-BIOINFO/CellDrift), a generalized linear model-based functional data analysis method that is capable of identifying covarying temporal patterns of various cell types in response to perturbations. As compared to several other approaches, CellDrift demonstrated superior performance in the identification of temporally varied perturbation patterns and the ability to impute missing time points. We applied CellDrift to multiple longitudinal datasets, including COVID-19 disease progression and gastrointestinal tract development, and demonstrated its ability to identify specific gene programs associated with sequential biological processes, trajectories and outcomes.

Clustered regularly interspaced short palindromic repeats (CRISPR)/Cas9 gene editing technology has been widely used to facilitate efficient genome editing. Current popular sgRNA design tools only consider the sgRNA perfectly matched to the target site and provide the results without any on-target mismatch. We suppose taking on-target gRNA-DNA mismatches into consideration might provide better sgRNA with similar binding activity and reduced off-target sites. Here, we trained a seq2seq-attention model with feedback-loop architecture, to automatically generate sgRNAs with on-target mismatches. Dual-luciferase reporter experiment showed that multiple sgRNAs with three mismatches could achieve the 80% of the relative activity of the perfect matched sgRNA. Meanwhile, it could reduce the number of off-target sites using sgRNAs with on-target mismatches. Finally, we provided a freely accessible web server sgRNA design tool named ExsgRNA. Users could submit their target sequence to this server and get optimal sgRNAs with less off-targets and similar on-target activity compared with the perfect-matched sgRNA.

Drug discovery is a lengthy, costly and high-risk endeavour that is further convoluted by high attrition rates in later development stages. Toxicity has been one of the main causes of failure during clinical trials, increasing drug development time and costs. To facilitate early identification and optimisation of toxicity profiles, several computational tools emerged aiming at improving success rates by timely pre-screening drug candidates. Despite these efforts, there is an increasing demand for platforms capable of assessing both environmental as well as human-based toxicity properties at large scale. Here, we present toxCSM, a comprehensive computational platform for the study and optimisation of toxicity profiles of small molecules. toxCSM leverages on the well-established concepts of graph-based signatures, molecular descriptors and similarity scores to develop 36 models for predicting a range of toxicity properties, which can assist in developing safer drugs and agrochemicals. toxCSM achieved an Area Under the Receiver Operating Characteristic (ROC) Curve (AUC) of up to 0.99 and Pearson's correlation coefficients of up to 0.94 on 10-fold cross-validation, with comparable performance on blind test sets, outperforming all alternative methods. toxCSM is freely available as a user-friendly web server and API at http://biosig.lab.uq.edu.au/toxcsm.

The identification of active binding drugs for target proteins (referred to as drug-target interaction prediction) is the key challenge in virtual screening, which plays an essential role in drug discovery. Although recent deep learning-based approaches achieve better performance than molecular docking, existing models often neglect topological or spatial of intermolecular information, hindering prediction performance. We recognize this problem and propose a novel approach called the Intermolecular Graph Transformer (IGT) that employs a dedicated attention mechanism to model intermolecular information with a three-way Transformer-based architecture. IGT outperforms state-of-the-art (SoTA) approaches by 9.1% and 20.5% over the second best option for binding activity and binding pose prediction, respectively, and exhibits superior generalization ability to unseen receptor proteins than SoTA approaches. Furthermore, IGT exhibits promising drug screening ability against severe acute respiratory syndrome coronavirus 2 by identifying 83.1% active drugs that have been validated by wet-lab experiments with near-native predicted binding poses. Source code and datasets are available at https://github.com/microsoft/IGT-Intermolecular-Graph-Transformer.

Liquid chromatography-mass spectrometry (LC-MS)-based untargeted metabolomics provides systematic profiling of metabolic. Yet, its applications in precision medicine (disease diagnosis) have been limited by several challenges, including metabolite identification, information loss and low reproducibility. Here, we present the deep-learning-based Pseudo-Mass Spectrometry Imaging (deepPseudoMSI) project (https://www.deeppseudomsi.org/), which converts LC-MS raw data to pseudo-MS images and then processes them by deep learning for precision medicine, such as disease diagnosis. Extensive tests based on real data demonstrated the superiority of deepPseudoMSI over traditional approaches and the capacity of our method to achieve an accurate individualized diagnosis. Our framework lays the foundation for future metabolic-based precision medicine.

We present a novel self-supervised Contrastive LEArning framework for single-cell ribonucleic acid (RNA)-sequencing (CLEAR) data representation and the downstream analysis. Compared with current methods, CLEAR overcomes the heterogeneity of the experimental data with a specifically designed representation learning task and thus can handle batch effects and dropout events simultaneously. It achieves superior performance on a broad range of fundamental tasks, including clustering, visualization, dropout correction, batch effect removal, and pseudo-time inference. The proposed method successfully identifies and illustrates inflammatory-related mechanisms in a COVID-19 disease study with 43 695 single cells from peripheral blood mononuclear cells.

Computational recovery of gene regulatory network (GRN) has recently undergone a great shift from bulk-cell towards designing algorithms targeting single-cell data. In this work, we investigate whether the widely available bulk-cell data could be leveraged to assist the GRN predictions for single cells. We infer cell-type-specific GRNs from both the single-cell RNA sequencing data and the generic GRN derived from the bulk cells by constructing a weakly supervised learning framework based on the axial transformer. We verify our assumption that the bulk-cell transcriptomic data are a valuable resource, which could improve the prediction of single-cell GRN by conducting extensive experiments. Our GRN-transformer achieves the state-of-the-art prediction accuracy in comparison to existing supervised and unsupervised approaches. In addition, we show that our method can identify important transcription factors and potential regulations for Alzheimer's disease risk genes by using the predicted GRN. Availability: The implementation of GRN-transformer is available at https://github.com/HantaoShu/GRN-Transformer.

Single-cell RNA-sequencing (scRNA-seq) has been widely used to depict gene expression profiles at the single-cell resolution. However, its relatively high dropout rate often results in artificial zero expressions of genes and therefore compromised reliability of results. To overcome such unwanted sparsity of scRNA-seq data, several imputation algorithms have been developed to recover the single-cell expression profiles. Here, we propose a novel approach, GE-Impute, to impute the dropout zeros in scRNA-seq data with graph embedding-based neural network model. GE-Impute learns the neural graph representation for each cell and reconstructs the cell-cell similarity network accordingly, which enables better imputation of dropout zeros based on the more accurately allocated neighbors in the similarity network. Gene expression correlation analysis between true expression data and simulated dropout data suggests significantly better performance of GE-Impute on recovering dropout zeros for both droplet- and plated-based scRNA-seq data. GE-Impute also outperforms other imputation methods in identifying differentially expressed genes and improving the unsupervised clustering on datasets from various scRNA-seq techniques. Moreover, GE-Impute enhances the identification of marker genes, facilitating the cell type assignment of clusters. In trajectory analysis, GE-Impute improves time-course scRNA-seq data analysis and reconstructing differentiation trajectory. The above results together demonstrate that GE-Impute could be a useful method to recover the single-cell expression profiles, thus enabling better biological interpretation of scRNA-seq data. GE-Impute is implemented in Python and is freely available at https://github.com/wxbCaterpillar/GE-Impute.

Recently, many studies have shown that lncRNA can mediate the regulation of TF-gene in drug sensitivity. However, there is still a lack of systematic identification of lncRNA-TF-gene regulatory triplets for drug sensitivity. In this study, we propose a novel analytic approach to systematically identify the lncRNA-TF-gene regulatory triplets related to the drug sensitivity by integrating transcriptome data and drug sensitivity data. Totally, 1570 drug sensitivity-related lncRNA-TF-gene triplets were identified, and 16 307 relationships were formed between drugs and triplets. Then, a comprehensive characterization was performed. Drug sensitivity-related triplets affect a variety of biological functions including drug response-related pathways. Phenotypic similarity analysis showed that the drugs with many shared triplets had high similarity in their two-dimensional structures and indications. In addition, Network analysis revealed the diverse regulation mechanism of lncRNAs in different drugs. Also, survival analysis indicated that lncRNA-TF-gene triplets related to the drug sensitivity could be candidate prognostic biomarkers for clinical applications. Next, using the random walk algorithm, the results of which we screen therapeutic drugs for patients across three cancer types showed high accuracy in the drug-cell line heterogeneity network based on the identified triplets. Besides, we developed a user-friendly web interface-DrugSETs (http://bio-bigdata.hrbmu.edu.cn/DrugSETs/) available to explore 1570 lncRNA-TF-gene triplets relevant with 282 drugs. It can also submit a patient's expression profile to predict therapeutic drugs conveniently. In summary, our research may promote the study of lncRNAs in the drug resistance mechanism and improve the effectiveness of treatment.

This letter points out a conceptual error made by the authors of a published paper, which presents a review and evaluation of computational methods in lncRNA identification. The error was made in the execution of the BASiNET method when considering an example file (toy model) made available by the authors with the aim of showing how a classification model could be stored in a file for later use. In this letter, this error is contextualized, the correct use of the BASiNET method is pointed out and the results of its correct execution to one of the datasets used in the review article are presented. The results clearly show the misuse of the method and present its correct use so that it can be fairly compared with other methods in the literature and prevent its misuse from being replicated by new studies.

Many biological applications are essentially pairwise comparison problems, such as evolutionary relationships on genomic sequences, contigs binning on metagenomic data, cell type identification on gene expression profiles of single-cells, etc. To make pair-wise comparison, it is necessary to adopt suitable dissimilarity metric. However, not all the metrics can be fully adapted to all possible biological applications. It is necessary to employ metric learning based on data adaptive to the application of interest. Therefore, in this study, we proposed MEtric Learning with Triplet network (MELT), which learns a nonlinear mapping from original space to the embedding space in order to keep similar data closer and dissimilar data far apart. MELT is a weakly supervised and data-driven comparison framework that offers more adaptive and accurate dissimilarity learned in the absence of the label information when the supervised methods are not applicable. We applied MELT in three typical applications of genomic data comparison, including hierarchical genomic sequences, longitudinal microbiome samples and longitudinal single-cell gene expression profiles, which have no distinctive grouping information. In the experiments, MELT demonstrated its empirical utility in comparison to many widely used dissimilarity metrics. And MELT is expected to accommodate a more extensive set of applications in large-scale genomic comparisons. MELT is available at https://github.com/Ying-Lab/MELT.

MOTIVATION: Accurate and efficient prediction of the molecular property is one of the fundamental problems in drug research and development. Recent advancements in representation learning have been shown to greatly improve the performance of molecular property prediction. However, due to limited labeled data, supervised learning-based molecular representation algorithms can only search limited chemical space and suffer from poor generalizability. RESULTS: In this work, we proposed a self-supervised learning method, ATMOL, for molecular representation learning and properties prediction. We developed a novel molecular graph augmentation strategy, referred to as attention-wise graph masking, to generate challenging positive samples for contrastive learning. We adopted the graph attention network as the molecular graph encoder, and leveraged the learned attention weights as masking guidance to generate molecular augmentation graphs. By minimization of the contrastive loss between original graph and augmented graph, our model can capture important molecular structure and higher order semantic information. Extensive experiments showed that our attention-wise graph mask contrastive learning exhibited state-of-the-art performance in a couple of downstream molecular property prediction tasks. We also verified that our model pretrained on larger scale of unlabeled data improved the generalization of learned molecular representation. Moreover, visualization of the attention heatmaps showed meaningful patterns indicative of atoms and atomic groups important to specific molecular property.

Precision medicine uses genetic, environmental and lifestyle factors to more accurately diagnose and treat disease in specific groups of patients, and it is considered one of the most promising medical efforts of our time. The use of genetics is arguably the most data-rich and complex components of precision medicine. The grand challenge today is the successful assimilation of genetics into precision medicine that translates across different ancestries, diverse diseases and other distinct populations, which will require clever use of artificial intelligence (AI) and machine learning (ML) methods. Our goal here was to review and compare scientific objectives, methodologies, datasets, data sources, ethics and gaps of AI/ML approaches used in genomics and precision medicine. We selected high-quality literature published within the last 5 years that were indexed and available through PubMed Central. Our scope was narrowed to articles that reported application of AI/ML algorithms for statistical and predictive analyses using whole genome and/or whole exome sequencing for gene variants, and RNA-seq and microarrays for gene expression. We did not limit our search to specific diseases or data sources. Based on the scope of our review and comparative analysis criteria, we identified 32 different AI/ML approaches applied in variable genomics studies and report widely adapted AI/ML algorithms for predictive diagnostics across several diseases.

The location of microRNAs (miRNAs) in cells determines their function in regulation activity. Studies have shown that miRNAs are stable in the extracellular environment that mediates cell-to-cell communication and are located in the intracellular region that responds to cellular stress and environmental stimuli. Though in situ detection techniques of miRNAs have made great contributions to the study of the localization and distribution of miRNAs, miRNA subcellular localization and their role are still in progress. Recently, some machine learning-based algorithms have been designed for miRNA subcellular location prediction, but their performance is still far from satisfactory. Here, we present a new data partitioning strategy that categorizes functionally similar locations for the precise and instructive prediction of miRNA subcellular location in Homo sapiens. To characterize the localization signals, we adopted one-hot encoding with post padding to represent the whole miRNA sequences, and proposed a deep bidirectional long short-term memory with the multi-head self-attention algorithm to model. The algorithm showed high selectivity in distinguishing extracellular miRNAs from intracellular miRNAs. Moreover, a series of motif analyses were performed to explore the mechanism of miRNA subcellular localization. To improve the convenience of the model, a user-friendly web server named iLoc-miRNA was established (http://iLoc-miRNA.lin-group.cn/).

Integration of single-cell transcriptome datasets from multiple sources plays an important role in investigating complex biological systems. The key to integration of transcriptome datasets is batch effect removal. Recent methods attempt to apply a contrastive learning strategy to correct batch effects. Despite their encouraging performance, the optimal contrastive learning framework for batch effect removal is still under exploration. We develop an improved contrastive learning-based batch correction framework, GLOBE. GLOBE defines adaptive translation transformations for each cell to guarantee the stability of approximating batch effects. To enhance the consistency of representations alignment, GLOBE utilizes a loss function that is both hardness-aware and consistency-aware to learn batch effect-invariant representations. Moreover, GLOBE computes batch-corrected gene matrix in a transparent approach to support diverse downstream analysis. Benchmarking results on a wide spectrum of datasets show that GLOBE outperforms other state-of-the-art methods in terms of robust batch mixing and superior conservation of biological signals. We further apply GLOBE to integrate two developing mouse neocortex datasets and show GLOBE succeeds in removing batch effects while preserving the contiguous structure of cells in raw data. Finally, a comprehensive study is conducted to validate the effectiveness of GLOBE.

Artificial intelligence (AI)-based computational techniques allow rapid exploration of the chemical space. However, representation of the compounds into computational-compatible and detailed features is one of the crucial steps for quantitative structure-activity relationship (QSAR) analysis. Recently, graph-based methods are emerging as a powerful alternative to chemistry-restricted fingerprints or descriptors for modeling. Although graph-based modeling offers multiple advantages, its implementation demands in-depth domain knowledge and programming skills. Here we introduce deepGraphh, an end-to-end web service featuring a conglomerate of established graph-based methods for model generation for classification or regression tasks. The graphical user interface of deepGraphh supports highly configurable parameter support for model parameter tuning, model generation, cross-validation and testing of the user-supplied query molecules. deepGraphh supports four widely adopted methods for QSAR analysis, namely, graph convolution network, graph attention network, directed acyclic graph and Attentive FP. Comparative analysis revealed that deepGraphh supported methods are comparable to the descriptors-based machine learning techniques. Finally, we used deepGraphh models to predict the blood-brain barrier permeability of human and microbiome-generated metabolites. In summary, deepGraphh offers a one-stop web service for graph-based methods for chemoinformatics.

In biomedical research, the replicability of findings across studies is highly desired. In this study, we focus on cancer omics data, for which the examination of replicability has been mostly focused on important omics variables identified in different studies. In published literature, although there have been extensive attention and ad hoc discussions, there is insufficient quantitative research looking into replicability measures and their properties. The goal of this study is to fill this important knowledge gap. In particular, we consider three sensible replicability measures, for which we examine distributional properties and develop a way of making inference. Applying them to three The Cancer Genome Atlas (TCGA) datasets reveals in general low replicability and significant across-data variations. To further comprehend such findings, we resort to simulation, which confirms the validity of the findings with the TCGA data and further informs the dependence of replicability on signal level (or equivalently sample size). Overall, this study can advance our understanding of replicability for cancer omics and other studies that have identification as a key goal.

Assessment of protective or harmful T cell response induced by any antigenic epitope is important in designing any immunotherapeutic molecule. The understanding of cytokine induction potential also helps us to monitor antigen-specific cellular immune responses and rational vaccine design. The classical immunoinformatics tools served well for prediction of B cell and T cell epitopes. However, in the last decade, the prediction algorithms for T cell epitope inducing specific cytokines have also been developed and appreciated in the scientific community. This review summarizes the current status of such tools, their applications, background algorithms, their use in experimental setup and functionalities available in the tools/web servers.

The authors of the BASiNET tool claim that the survey paper 'A systematic evaluation of computational tools for lncRNA identification' incorrectly evaluates the BASiNET tool. Here, we point out that the survey paper correctly evaluates the BASiNET tool and why the evaluation should not be carried out as BASiNET authors suggest.

Genetic association studies have been very successful at elucidating the genetic background of many complex diseases/traits. However, the X-chromosome is often neglected in these studies because of technical difficulties and the fact that most tools only utilize genetic data from autosomes. In this review, we aim to provide an overview of different practical approaches that are followed to incorporate the X-chromosome in association analysis, such as Genome-Wide Association Studies and Expression Quantitative Trait Loci Analysis. In general, the choice of which test statistics is most appropriate will depend on three main criteria: (1) the underlying X-inactivation model, (2) if Hardy-Weinberg equilibrium holds and sex-specific allele frequencies are expected and (3) whether adjustment for confounding variables is required. All in all, it is recommended that a combination of different association tests should be used for the analysis of X-chromosome.

Proteins/peptides have shown to be promising therapeutic agents for a variety of diseases. However, toxicity is one of the obstacles in protein/peptide-based therapy. The current study describes a web-based tool, ToxinPred2, developed for predicting the toxicity of proteins. This is an update of ToxinPred developed mainly for predicting toxicity of peptides and small proteins. The method has been trained, tested and evaluated on three datasets curated from the recent release of the SwissProt. To provide unbiased evaluation, we performed internal validation on 80% of the data and external validation on the remaining 20% of data. We have implemented the following techniques for predicting protein toxicity; (i) Basic Local Alignment Search Tool-based similarity, (ii) Motif-EmeRging and with Classes-Identification-based motif search and (iii) Prediction models. Similarity and motif-based techniques achieved a high probability of correct prediction with poor sensitivity/coverage, whereas models based on machine-learning techniques achieved balance sensitivity and specificity with reasonably high accuracy. Finally, we developed a hybrid method that combined all three approaches and achieved a maximum area under receiver operating characteristic curve around 0.99 with Matthews correlation coefficient 0.91 on the validation dataset. In addition, we developed models on alternate and realistic datasets. The best machine learning models have been implemented in the web server named 'ToxinPred2', which is available at https://webs.iiitd.edu.in/raghava/toxinpred2/ and a standalone version at https://github.com/raghavagps/toxinpred2. This is a general method developed for predicting the toxicity of proteins regardless of their source of origin.

Since the problem proposed in late 2000s, microRNA-disease association (MDA) predictions have been implemented based on the data fusion paradigm. Integrating diverse data sources gains a more comprehensive research perspective, and brings a challenge to algorithm design for generating accurate, concise and consistent representations of the fused data. After more than a decade of research progress, a relatively simple algorithm like the score function or a single computation layer may no longer be sufficient for further improving predictive performance. Advanced model design has become more frequent in recent years, particularly in the form of reasonably combing multiple algorithms, a process known as model fusion. In the current review, we present 29 state-of-the-art models and introduce the taxonomy of computational models for MDA prediction based on model fusion and non-fusion. The new taxonomy exhibits notable changes in the algorithmic architecture of models, compared with that of earlier ones in the 2017 review by Chen et al. Moreover, we discuss the progresses that have been made towards overcoming the obstacles to effective MDA prediction since 2017 and elaborated on how future models can be designed according to a set of new schemas. Lastly, we analysed the strengths and weaknesses of each model category in the proposed taxonomy and proposed future research directions from diverse perspectives for enhancing model performance.

Circular RNAs (circRNAs) are involved in the regulatory mechanisms of multiple complex diseases, and the identification of their associations is critical to the diagnosis and treatment of diseases. In recent years, many computational methods have been designed to predict circRNA-disease associations. However, most of the existing methods rely on single correlation data. Here, we propose a machine learning framework for circRNA-disease association prediction, called MLCDA, which effectively fuses multiple sources of heterogeneous information including circRNA sequences and disease ontology. Comprehensive evaluation in the gold standard dataset showed that MLCDA can successfully capture the complex relationships between circRNAs and diseases and accurately predict their potential associations. In addition, the results of case studies on real data show that MLCDA significantly outperforms other existing methods. MLCDA can serve as a useful tool for circRNA-disease association prediction, providing mechanistic insights for disease research and thus facilitating the progress of disease treatment.

MOTIVATION: Digital pathological analysis is run as the main examination used for cancer diagnosis. Recently, deep learning-driven feature extraction from pathology images is able to detect genetic variations and tumor environment, but few studies focus on differential gene expression in tumor cells. RESULTS: In this paper, we propose a self-supervised contrastive learning framework, HistCode, to infer differential gene expression from whole slide images (WSIs). We leveraged contrastive learning on large-scale unannotated WSIs to derive slide-level histopathological features in latent space, and then transfer it to tumor diagnosis and prediction of differentially expressed cancer driver genes. Our experiments showed that our method outperformed other state-of-the-art models in tumor diagnosis tasks, and also effectively predicted differential gene expression. Interestingly, we found the genes with higher fold change can be more precisely predicted. To intuitively illustrate the ability to extract informative features from pathological images, we spatially visualized the WSIs colored by the attention scores of image tiles. We found that the tumor and necrosis areas were highly consistent with the annotations of experienced pathologists. Moreover, the spatial heatmap generated by lymphocyte-specific gene expression patterns was also consistent with the manually labeled WSIs.

Single-cell sequencing technologies are widely used to discover the evolutionary relationships and the differences in cells. Since dropout events may frustrate the analysis, many imputation approaches for single-cell RNA-seq data have appeared in previous attempts. However, previous imputation attempts usually suffer from the over-smooth problem, which may bring limited improvement or negative effect for the downstream analysis of single-cell RNA-seq data. To solve this difficulty, we propose a novel two-stage diffusion-denoising method called SCDD for large-scale single-cell RNA-seq imputation in this paper. We introduce the diffusion i.e. a direct imputation strategy using the expression of similar cells for potential dropout sites, to perform the initial imputation at first. After the diffusion, a joint model integrated with graph convolutional neural network and contractive autoencoder is developed to generate superposition states of similar cells, from which we restore the original states and remove the noise introduced by the diffusion. The final experimental results indicate that SCDD could effectively suppress the over-smooth problem and remarkably improve the effect of single-cell RNA-seq downstream analysis, including clustering and trajectory analysis.

The quantification of developmental potential is critical for determining developmental stages and identifying essential molecular signatures in single-cell studies. Here, we present FitDevo, a novel method for inferring developmental potential using scRNA-seq data. The main idea of FitDevo is first to generate sample-specific gene weight (SSGW) and then infer developmental potential by calculating the correlation between SSGW and gene expression. SSGW is generated using a generalized linear model that combines sample-specific information and gene weight learned from a training dataset covering scRNA-seq data of 17 previously published datasets. We have rigorously validated FitDevo's effectiveness using a testing dataset with scRNA-seq data from 28 existing datasets and have also demonstrated its superiority over current methods. Furthermore, FitDevo's broad application scope has been illustrated using three practical scenarios: deconvolution analysis of epidermis, spatial transcriptomic data analysis of hearts and intestines, and developmental potential analysis of breast cancer. The source code and related data are available at https://github.com/jumphone/fitdevo.

Deoxyribonucleic acid (DNA)-based data storage is a promising new storage technology which has the advantage of high storage capacity and long storage time compared with traditional storage media. However, the synthesis and sequencing process of DNA can randomly generate many types of errors, which makes it more difficult to cluster DNA sequences to recover DNA information. Currently, the available DNA clustering algorithms are targeted at DNA sequences in the biological domain, which not only cannot adapt to the characteristics of sequences in DNA storage, but also tend to be unacceptably time-consuming for billions of DNA sequences in DNA storage. In this paper, we propose an efficient DNA clustering method termed Clover for DNA storage with linear computational complexity and low memory. Clover avoids the computation of the Levenshtein distance by using a tree structure for interval-specific retrieval. We argue through theoretical proofs that Clover has standard linear computational complexity, low space complexity, etc. Experiments show that our method can cluster 10 million DNA sequences into 50 000 classes in 10 s and meet an accuracy rate of over 99%. Furthermore, we have successfully completed an unprecedented clustering of 10 billion DNA data on a single home computer and the time consumption still satisfies the linear relationship. Clover is freely available at https://github.com/Guanjinqu/Clover.

With >1 million associated deaths in 2020, human tuberculosis (TB) caused by the bacteria Mycobacterium tuberculosis remains one of the deadliest infectious diseases. A plethora of genomic tools and bioinformatics pipelines have become available in recent years to assist the whole genome sequencing of M. tuberculosis. The Oxford Nanopore Technologies (ONT) portable sequencer is a promising platform for cost-effective application in clinics, including personalizing treatment through detection of drug resistance-associated mutations, or in the field, to assist epidemiological and transmission investigations. In this study, we performed a comparison of 10 clinical isolates with DNA sequenced on both long-read ONT and (gold standard) short-read Illumina HiSeq platforms. Our analysis demonstrates the robustness of the ONT variant calling for single nucleotide polymorphisms, despite the high error rate. Moreover, because of improved coverage in repetitive regions where short sequencing reads fail to align accurately, ONT data analysis can incorporate additional regions of the genome usually excluded (e.g. pe/ppe genes). The resulting extra resolution can improve the characterization of transmission clusters and dynamics based on inferring closely related isolates. High concordance in variants in loci associated with drug resistance supports its use for the rapid detection of resistant mutations. Overall, ONT sequencing is a promising tool for TB genomic investigations, particularly to inform clinical and surveillance decision-making to reduce the disease burden.

Many chemicals are present in our environment, and all living species are exposed to them. However, numerous chemicals pose risks, such as developing severe diseases, if they occur at the wrong time in the wrong place. For the majority of the chemicals, these risks are not known. Chemical risk assessment and subsequent regulation of use require efficient and systematic strategies. Lab-based methods-even if high throughput-are too slow to keep up with the pace of chemical innovation. Existing computational approaches are designed for specific chemical classes or sub-problems but not usable on a large scale. Further, the application range of these approaches is limited by the low amount of available labeled training data. We present the ready-to-use and stand-alone program deepFPlearn that predicts the association between chemical structures and effects on the gene/pathway level using a combined deep learning approach. deepFPlearn uses a deep autoencoder for feature reduction before training a deep feed-forward neural network to predict the target association. We received good prediction qualities and showed that our feature compression preserves relevant chemical structural information. Using a vast chemical inventory (unlabeled data) as input for the autoencoder did not reduce our prediction quality but allowed capturing a much more comprehensive range of chemical structures. We predict meaningful-experimentally verified-associations of chemicals and effects on unseen data. deepFPlearn classifies hundreds of thousands of chemicals in seconds. We provide deepFPlearn as an open-source and flexible tool that can be easily retrained and customized to different application settings at https://github.com/yigbt/deepFPlearn.

Clustering analysis is widely used in single-cell ribonucleic acid (RNA)-sequencing (scRNA-seq) data to discover cell heterogeneity and cell states. While many clustering methods have been developed for scRNA-seq analysis, most of these methods require to provide the number of clusters. However, it is not easy to know the exact number of cell types in advance, and experienced determination is not always reliable. Here, we have developed ADClust, an automatic deep embedding clustering method for scRNA-seq data, which can accurately cluster cells without requiring a predefined number of clusters. Specifically, ADClust first obtains low-dimensional representation through pre-trained autoencoder and uses the representations to cluster cells into initial micro-clusters. The clusters are then compared in between by a statistical test, and similar micro-clusters are merged into larger clusters. According to the clustering, cell representations are updated so that each cell will be pulled toward centers of its assigned cluster and similar clusters, while cells are separated to keep distances between clusters. This is accomplished through jointly optimizing the carefully designed clustering and autoencoder loss functions. This merging process continues until convergence. ADClust was tested on 11 real scRNA-seq datasets and was shown to outperform existing methods in terms of both clustering performance and the accuracy on the number of the determined clusters. More importantly, our model provides high speed and scalability for large datasets.

Increasing evidence has suggested that microRNAs (miRNAs) are important biomarkers of various diseases. Numerous graph neural network (GNN) models have been proposed for predicting miRNA-disease associations. However, the existing GNN-based methods have over-smoothing issue-the learned feature embeddings of miRNA nodes and disease nodes are indistinguishable when stacking multiple GNN layers. This issue makes the performance of the methods sensitive to the number of layers, and significantly hurts the performance when more layers are employed. In this study, we resolve this issue by a novel self-feature-based graph autoencoder model, shortened as SFGAE. The key novelty of SFGAE is to construct miRNA-self embeddings and disease-self embeddings, and let them be independent of graph interactions between two types of nodes. The novel self-feature embeddings enrich the information of typical aggregated feature embeddings, which aggregate the information from direct neighbors and hence heavily rely on graph interactions. SFGAE adopts a graph encoder with attention mechanism to concatenate aggregated feature embeddings and self-feature embeddings, and adopts a bilinear decoder to predict links. Our experiments show that SFGAE achieves state-of-the-art performance. In particular, SFGAE improves the average AUC upon recent GAEMDA [1] on the benchmark datasets HMDD v2.0 and HMDD v3.2, and consistently performs better when less (e.g. 10%) training samples are used. Furthermore, SFGAE effectively overcomes the over-smoothing issue and performs stably well on deeper models (e.g. eight layers). Finally, we carry out case studies on three human diseases, colon neoplasms, esophageal neoplasms and kidney neoplasms, and perform a survival analysis using kidney neoplasm as an example. The results suggest that SFGAE is a reliable tool for predicting potential miRNA-disease associations.

MOTIVATION: The associations between biomarkers and human diseases play a key role in understanding complex pathology and developing targeted therapies. Wet lab experiments for biomarker discovery are costly, laborious and time-consuming. Computational prediction methods can be used to greatly expedite the identification of candidate biomarkers. RESULTS: Here, we present a novel computational model named GTGenie for predicting the biomarker-disease associations based on graph and text features. In GTGenie, a graph attention network is utilized to characterize diverse similarities of biomarkers and diseases from heterogeneous information resources. Meanwhile, a pretrained BERT-based model is applied to learn the text-based representation of biomarker-disease relation from biomedical literature. The captured graph and text features are then integrated in a bimodal fusion network to model the hybrid entity representation. Finally, inductive matrix completion is adopted to infer the missing entries for reconstructing relation matrix, with which the unknown biomarker-disease associations are predicted. Experimental results on HMDD, HMDAD and LncRNADisease data sets showed that GTGenie can obtain competitive prediction performance with other state-of-the-art methods. AVAILABILITY: The source code of GTGenie and the test data are available at: https://github.com/Wolverinerine/GTGenie.

Tipping points or critical transitions widely exist during the progression of many biological processes. It is of great importance to detect the tipping point with the measured omics data, which may be a key to achieving predictive or preventive medicine. We present the tipping point detector (TPD), a web tool for the detection of the tipping point during the dynamic process of biological systems, and further its leading molecules or network, based on the input high-dimensional time series or stage course data. With the solid theoretical background of dynamic network biomarker (DNB) and a series of computational methods for DNB detection, TPD detects the potential tipping point/critical state from the input omics data and outputs multifarious visualized results, including a suggested tipping point with a statistically significant P value, the identified key genes and their functional biological information, the dynamic change in the DNB/leading network that may drive the critical transition and the survival analysis based on DNB scores that may help to identify 'dark' genes (nondifferential in terms of expression but differential in terms of DNB scores). TPD fits all current browsers, such as Chrome, Firefox, Edge, Opera, Safari and Internet Explorer. TPD is freely accessible at http://www.rpcomputationalbiology.cn/TPD.

Somatic variants act as critical players during cancer occurrence and development. Thus, an accurate and robust method to identify them is the foundation of cutting-edge cancer genome research. However, due to low accessibility and high individual-/sample-specificity of the somatic variants in tumor samples, the detection is, to date, still crammed with challenges, particularly when lacking paired normal samples as control. To solve this burning issue, we developed a tumor-only somatic and germline variant identification method (TSomVar) using the random forest algorithm established on sample-specific variant datasets derived from genotype imputation, reads-mapping level annotation and functional annotation. We trained TSomVar by using genomic variant datasets of three major cancer types: colorectal cancer, hepatocellular carcinoma and skin cutaneous melanoma. Compared with existing tumor-only somatic variant identification tools, TSomVar shows excellent performances in somatic variant detection with higher accuracy and better capability of recalling for test datasets from colorectal cancer and skin cutaneous melanoma. In addition, TSomVar is equipped with the competence of accurately identifying germline variants in tumor samples. Taken together, TSomVar will undoubtedly facilitate and revolutionize somatic variant explorations in cancer research.

Nowadays, the complexity of disease mechanisms and the inadequacy of single-target therapies in restoring the biological system have inevitably instigated the strategy of multi-target therapeutics with the analysis of each target individually. However, it is not suitable for dealing with the conflicts between targets or between drugs. With the release of high-precision protein structure prediction artificial intelligence, large-scale high-precision protein structure prediction and docking have become possible. In this article, we propose a multi-target drug discovery method by the example of therapeutic hypothermia (TH). First, we performed protein structure prediction for all protein targets of each group by AlphaFold2 and RoseTTAFold. Then, QuickVina 2 is used for molecular docking between the proteins and drugs. After docking, we use PageRank to rank single drugs and drug combinations of each group. The ePharmaLib was used for predicting the side effect targets. Given the differences in the weights of different targets, the method can effectively avoid inhibiting beneficial proteins while inhibiting harmful proteins. So it could minimize the conflicts between different doses and be friendly to chronotherapeutics. Besides, this method also has potential in precision medicine for its high compatibility with bioinformatics and promotes the development of pharmacogenomics and bioinfo-pharmacology.

Drug combination therapies are superior to monotherapy for cancer treatment in many ways. Identifying novel drug combinations by screening is challenging for the wet-lab experiments due to the time-consuming process of the enormous search space of possible drug pairs. Thus, computational methods have been developed to predict drug pairs with potential synergistic functions. Notwithstanding the success of current models, understanding the mechanism of drug synergy from a chemical-gene-tissue interaction perspective lacks study, hindering current algorithms from drug mechanism study. Here, we proposed a deep neural network model termed DTSyn (Dual Transformer encoder model for drug pair Synergy prediction) based on a multi-head attention mechanism to identify novel drug combinations. We designed a fine-granularity transformer encoder to capture chemical substructure-gene and gene-gene associations and a coarse-granularity transformer encoder to extract chemical-chemical and chemical-cell line interactions. DTSyn achieved the highest receiver operating characteristic area under the curve of 0.73, 0.78. 0.82 and 0.81 on four different cross-validation tasks, outperforming all competing methods. Further, DTSyn achieved the best True Positive Rate (TPR) over five independent data sets. The ablation study showed that both transformer encoder blocks contributed to the performance of DTSyn. In addition, DTSyn can extract interactions among chemicals and cell lines, representing the potential mechanisms of drug action. By leveraging the attention mechanism and pretrained gene embeddings, DTSyn shows improved interpretability ability. Thus, we envision our model as a valuable tool to prioritize synergistic drug pairs with chemical and cell line gene expression profile.

Neuropeptides (NPs) are a particular class of informative substances in the immune system and physiological regulation. They play a crucial role in regulating physiological functions in various biological growth and developmental stages. In addition, NPs are crucial for developing new drugs for the treatment of neurological diseases. With the development of molecular biology techniques, some data-driven tools have emerged to predict NPs. However, it is necessary to improve the predictive performance of these tools for NPs. In this study, we developed a deep learning model (NeuroPred-CLQ) based on the temporal convolutional network (TCN) and multi-head attention mechanism to identify NPs effectively and translate the internal relationships of peptide sequences into numerical features by the Word2vec algorithm. The experimental results show that NeuroPred-CLQ learns data information effectively, achieving 93.6% accuracy and 98.8% AUC on the independent test set. The model has better performance in identifying NPs than the state-of-the-art predictors. Visualization of features using t-distribution random neighbor embedding shows that the NeuroPred-CLQ can clearly distinguish the positive NPs from the negative ones. We believe the NeuroPred-CLQ can facilitate drug development and clinical trial studies to treat neurological disorders.

Circular RNAs (circRNAs) are a class of structurally stable endogenous noncoding RNA molecules. Increasing studies indicate that circRNAs play vital roles in human diseases. However, validating disease-related circRNAs in vivo is costly and time-consuming. A reliable and effective computational method to identify circRNA-disease associations deserves further studies. In this study, we propose a computational method called RNMFLP that combines robust nonnegative matrix factorization (RNMF) and label propagation algorithm (LP) to predict circRNA-disease associations. First, to reduce the impact of false negative data, the original circRNA-disease adjacency matrix is updated by matrix multiplication using the integrated circRNA similarity and the disease similarity information. Subsequently, the RNMF algorithm is used to obtain the restricted latent space to capture potential circRNA-disease pairs from the association matrix. Finally, the LP algorithm is utilized to predict more accurate circRNA-disease associations from the integrated circRNA similarity network and integrated disease similarity network, respectively. Fivefold cross-validation of four datasets shows that RNMFLP is superior to the state-of-the-art methods. In addition, case studies on lung cancer, hepatocellular carcinoma and colorectal cancer further demonstrate the reliability of our method to discover disease-related circRNAs.

Cell surface proteins have been used as diagnostic and prognostic markers in cancer research and as targets for the development of anticancer agents. Many of these proteins lie at the top of signaling cascades regulating cell responses and gene expression, therefore acting as 'signaling hubs'. It has been previously demonstrated that the integrated network analysis on transcriptomic data is able to infer cell surface protein activity in breast cancer. Such an approach has been implemented in a publicly available method called 'SURFACER'. SURFACER implements a network-based analysis of transcriptomic data focusing on the overall activity of curated surface proteins, with the final aim to identify those proteins driving major phenotypic changes at a network level, named surface signaling hubs. Here, we show the ability of SURFACER to discover relevant knowledge within and across cancer datasets. We also show how different cancers can be stratified in surface-activity-specific groups. Our strategy may identify cancer-wide markers to design targeted therapies and biomarker-based diagnostic approaches.

An unsolved challenge in developing molecular representation is determining an optimal method to characterize the molecular structure. Comprehension of intramolecular interactions is paramount toward achieving this goal. In this study, ComABAN, a new graph-attention-based approach, is proposed to improve the accuracy of molecular representation by simultaneously considering atom-atom, bond-bond and atom-bond interactions. In addition, we benchmark models extensively on 8 public and 680 proprietary industrial datasets spanning a wide variety of chemical end points. The results show that ComABAN has higher prediction accuracy compared with the classical machine learning method and the deep learning-based methods. Furthermore, the trained neural network was used to predict a library of 1.5 million molecules and picked out compounds with a classification result of grade I. Subsequently, these predicted molecules were scored and ranked using cascade docking, molecular dynamics simulations to generate five potential candidates. All five molecules showed high similarity to nanomolar bioactive inhibitors suppressing the expression of HIF-1α, and we synthesized three compounds (Y-1, Y-3, Y-4) and tested their inhibitory ability in vitro. Our results indicate that ComABAN is an effective tool for accelerating drug discovery.

CRISPR-Cas system is an adaptive immune system widely found in most bacteria and archaea to defend against exogenous gene invasion. One of the most critical steps in the study of exploring and classifying novel CRISPR-Cas systems and their functional diversity is the identification of Cas proteins in CRISPR-Cas systems. The discovery of novel Cas proteins has also laid the foundation for technologies such as CRISPR-Cas-based gene editing and gene therapy. Currently, accurate and efficient screening of Cas proteins from metagenomic sequences and proteomic sequences remains a challenge. For Cas proteins with low sequence conservation, existing tools for Cas protein identification based on homology cannot guarantee identification accuracy and efficiency. In this paper, we have developed a novel stacking-based ensemble learning framework for Cas protein identification, called CRISPRCasStack. In particular, we applied the SHAP (SHapley Additive exPlanations) method to analyze the features used in CRISPRCasStack. Sufficient experimental validation and independent testing have demonstrated that CRISPRCasStack can address the accuracy deficiencies and inefficiencies of the existing state-of-the-art tools. We also provide a toolkit to accurately identify and analyze potential Cas proteins, Cas operons, CRISPR arrays and CRISPR-Cas locus in prokaryotic sequences. The CRISPRCasStack toolkit is available at https://github.com/yrjia1015/CRISPRCasStack.

Protein S-sulfinylation is an important posttranslational modification that regulates a variety of cell and protein functions. This modification has been linked to signal transduction, redox homeostasis and neuronal transmission in studies. Therefore, identification of S-sulfinylation sites is crucial to understanding its structure and function, which is critical in cell biology and human diseases. In this study, we propose a multi-module deep learning framework named DLF-Sul for identification of S-sulfinylation sites in proteins. First, three types of features are extracted including binary encoding, BLOSUM62 and amino acid index. Then, sequential features are further extracted based on these three types of features using bidirectional long short-term memory network. Next, multi-head self-attention mechanism is utilized to filter the effective attribute information, and residual connection helps to reduce information loss. Furthermore, convolutional neural network is employed to extract local deep features information. Finally, fully connected layers acts as classifier that map samples to corresponding label. Performance metrics on independent test set, including sensitivity, specificity, accuracy, Matthews correlation coefficient and area under curve, reach 91.80%, 92.36%, 92.08%, 0.8416 and 96.40%, respectively. The results show that DLF-Sul is an effective tool for predicting S-sulfinylation sites. The source code is available on the website https://github.com/ningq669/DLF-Sul.

The rapid development of single-cel+l RNA sequencing (scRNA-seq) technology provides unprecedented opportunities for exploring biological phenomena at the single-cell level. The discovery of cell types is one of the major applications for researchers to explore the heterogeneity of cells. Some computational methods have been proposed to solve the problem of scRNA-seq data clustering. However, the unavoidable technical noise and notorious dropouts also reduce the accuracy of clustering methods. Here, we propose the cauchy-based bounded constraint low-rank representation (CBLRR), which is a low-rank representation-based method by introducing cauchy loss function (CLF) and bounded nuclear norm regulation, aiming to alleviate the above issue. Specifically, as an effective loss function, the CLF is proven to enhance the robustness of the identification of cell types. Then, we adopt the bounded constraint to ensure the entry values of single-cell data within the restricted interval. Finally, the performance of CBLRR is evaluated on 15 scRNA-seq datasets, and compared with other state-of-the-art methods. The experimental results demonstrate that CBLRR performs accurately and robustly on clustering scRNA-seq data. Furthermore, CBLRR is an effective tool to cluster cells, and provides great potential for downstream analysis of single-cell data. The source code of CBLRR is available online at https://github.com/Ginnay/CBLRR.

Human leukocyte antigens (HLA) regulate various innate and adaptive immune responses and play a crucial immunomodulatory role. Recent studies revealed that non-classical HLA-(HLA-E & HLA-G) based immunotherapies have many advantages over traditional HLA-based immunotherapy, particularly against cancer and COVID-19 infection. In the last two decades, several methods have been developed to predict the binders of classical HLA alleles. In contrast, limited attempts have been made to develop methods for predicting non-classical HLA binding peptides, due to the scarcity of sufficient experimental data. Of note, in order to facilitate the scientific community, we have developed an artificial intelligence-based method for predicting binders of class-Ib HLA alleles. All the models were trained and tested on experimentally validated data obtained from the recent release of IEDB. The machine learning models achieved more than 0.98 AUC for HLA-G alleles on validation dataset. Similarly, our models achieved the highest AUC of 0.96 and 0.94 on the validation dataset for HLA-E*01:01 and HLA-E*01:03, respectively. We have summarized the models developed in the past for non-classical HLA and validated the performance with the models developed in this study. Moreover, to facilitate the community, we have utilized our tool for predicting the potential non-classical HLA binding peptides in the spike protein of different variants of virus causing COVID-19, including Omicron (B.1.1.529). One of the major challenges in the field of immunotherapy is to identify the promiscuous binders or antigenic regions that can bind to a large number of HLA alleles. To predict the promiscuous binders for the non-classical HLA alleles, we developed a web server HLAncPred (https://webs.iiitd.edu.in/raghava/hlancpred) and standalone package.

Measuring the semantic similarity between Gene Ontology (GO) terms is a fundamental step in numerous functional bioinformatics applications. To fully exploit the metadata of GO terms, word embedding-based methods have been proposed recently to map GO terms to low-dimensional feature vectors. However, these representation methods commonly overlook the key information hidden in the whole GO structure and the relationship between GO terms. In this paper, we propose a novel representation model for GO terms, named GT2Vec, which jointly considers the GO graph structure obtained by graph contrastive learning and the semantic description of GO terms based on BERT encoders. Our method is evaluated on a protein similarity task on a collection of benchmark datasets. The experimental results demonstrate the effectiveness of using a joint encoding graph structure and textual node descriptors to learn vector representations for GO terms.

Statistical and machine learning techniques based on relative abundances have been used to predict health conditions and to identify microbial biomarkers. However, high dimensionality, sparsity and the compositional nature of microbiome data represent statistical challenges. On the other hand, the taxon grouping allows summarizing microbiome abundance with a coarser resolution in a lower dimension, but it presents new challenges when correlating taxa with a disease. In this work, we present a novel approach that groups Operational Taxonomical Units (OTUs) based only on relative abundances as an alternative to taxon grouping. The proposed procedure acknowledges the compositional data making use of principal balances. The identified groups are called Principal Microbial Groups (PMGs). The procedure reduces the need for user-defined aggregation of $\textrm{OTU}$s and offers the possibility of working with coarse group of $\textrm{OTU}$s, which are not present in a phylogenetic tree. PMGs can be used for two different goals: (1) as a dimensionality reduction method for compositional data, (2) as an aggregation procedure that provides an alternative to taxon grouping for construction of microbial balances afterward used for disease prediction. We illustrate the procedure with a cirrhosis study data. PMGs provide a coherent data analysis for the search of biomarkers in human microbiota. The source code and demo data for PMGs are available at: https://github.com/asliboyraz/PMGs.

Accurate inference of gene regulatory networks (GRNs) is an essential premise for understanding pathogenesis and curing diseases. Various computational methods have been developed for GRN inference, but the identification of redundant regulation remains a challenge faced by researchers. Although combining global and local topology can identify and reduce redundant regulations, the topologies' specific forms and cooperation modes are unclear and real regulations may be sacrificed. Here, we propose a network structure control method [network-structure-controlling-based GRN inference method (NSCGRN)] that stipulates the global and local topology's specific forms and cooperation mode. The method is carried out in a cooperative mode of 'global topology dominates and local topology refines'. Global topology requires layering and sparseness of the network, and local topology requires consistency of the subgraph association pattern with the network motifs (fan-in, fan-out, cascade and feedforward loop). Specifically, an ordered gene list is obtained by network topology centrality sorting. A Bernaola-Galvan mutation detection algorithm applied to the list gives the hierarchy of GRNs to control the upstream and downstream regulations within the global scope. Finally, four network motifs are integrated into the hierarchy to optimize local complex regulations and form a cooperative mode where global and local topologies play the dominant and refined roles, respectively. NSCGRN is compared with state-of-the-art methods on three different datasets (six networks in total), and it achieves the highest F1 and Matthews correlation coefficient. Experimental results show its unique advantages in GRN inference.

The discovery of drug-target interactions (DTIs) is a very promising area of research with great potential. The accurate identification of reliable interactions among drugs and proteins via computational methods, which typically leverage heterogeneous information retrieved from diverse data sources, can boost the development of effective pharmaceuticals. Although random walk and matrix factorization techniques are widely used in DTI prediction, they have several limitations. Random walk-based embedding generation is usually conducted in an unsupervised manner, while the linear similarity combination in matrix factorization distorts individual insights offered by different views. To tackle these issues, we take a multi-layered network approach to handle diverse drug and target similarities, and propose a novel optimization framework, called Multiple similarity DeepWalk-based Matrix Factorization (MDMF), for DTI prediction. The framework unifies embedding generation and interaction prediction, learning vector representations of drugs and targets that not only retain higher order proximity across all hyper-layers and layer-specific local invariance, but also approximate the interactions with their inner product. Furthermore, we develop an ensemble method (MDMF2A) that integrates two instantiations of the MDMF model, optimizing the area under the precision-recall curve (AUPR) and the area under the receiver operating characteristic curve (AUC), respectively. The empirical study on real-world DTI datasets shows that our method achieves statistically significant improvement over current state-of-the-art approaches in four different settings. Moreover, the validation of highly ranked non-interacting pairs also demonstrates the potential of MDMF2A to discover novel DTIs.

Anatomical Therapeutic Chemical (ATC) classification for compounds/drugs plays an important role in drug development and basic research. However, previous methods depend on interactions extracted from STITCH dataset which may make it depend on lab experiments. We present a pilot study to explore the possibility of conducting the ATC prediction solely based on the molecular structures. The motivation is to eliminate the reliance on the costly lab experiments so that the characteristics of a drug can be pre-assessed for better decision-making and effort-saving before the actual development. To this end, we construct a new benchmark consisting of 4545 compounds which is with larger scale than the one used in previous study. A light-weight prediction model is proposed. The model is with better explainability in the sense that it is consists of a straightforward tokenization that extracts and embeds statistically and physicochemically meaningful tokens, and a deep network backed by a set of pyramid kernels to capture multi-resolution chemical structural characteristics. Its efficacy has been validated in the experiments where it outperforms the state-of-the-art methods by 15.53% in accuracy and by 69.66% in terms of efficiency. We make the benchmark dataset, source code and web server open to ease the reproduction of this study.

Nonspecific cross-linker can provide distance restraints between surface residues of any type, which could be used to investigate protein structure construction and protein-protein interaction (PPI). However, the vast number of potential combinations of cross-linked residues or sites obtained with such a cross-linker makes the data challenging to analyze, especially for the proteome-wide applications. Here, we developed SpotLink software for identifying site nonspecific cross-links at the proteome scale. Contributed by the dual pointer dynamic pruning algorithm and the quality control of cross-linking sites, SpotLink identified > 3000 cross-links from human cell samples within a short period of days. We demonstrated that SpotLink outperformed other approaches in terms of sensitivity and precision on the datasets of the simulated succinimidyl 4,4'-azipentanoate dataset and the condensin complexes with known structures. In addition, some valuable PPI were discovered in the datasets of the condensin complexes and the HeLa dataset, indicating the unique identification advantages of site nonspecific cross-linking. These findings reinforce the importance of SpotLink as a fundamental characteristic of site nonspecific cross-linking technologies.

Accurate identification of genetic variants from family child-mother-father trio sequencing data is important in genomics. However, state-of-the-art approaches treat variant calling from trios as three independent tasks, which limits their calling accuracy for Nanopore long-read sequencing data. For better trio variant calling, we introduce Clair3-Trio, the first variant caller tailored for family trio data from Nanopore long-reads. Clair3-Trio employs a Trio-to-Trio deep neural network model, which allows it to input the trio sequencing information and output all of the trio's predicted variants within a single model to improve variant calling. We also present MCVLoss, a novel loss function tailor-made for variant calling in trios, leveraging the explicit encoding of the Mendelian inheritance. Clair3-Trio showed comprehensive improvement in experiments. It predicted far fewer Mendelian inheritance violation variations than current state-of-the-art methods. We also demonstrated that our Trio-to-Trio model is more accurate than competing architectures. Clair3-Trio is accessible as a free, open-source project at https://github.com/HKU-BAL/Clair3-Trio.

DNA-protein interaction is one of the most crucial interactions in the biological system, which decides the fate of many processes such as transcription, regulation and splicing of genes. In this study, we trained our models on a training dataset of 646 DNA-binding proteins having 15 636 DNA interacting and 298 503 non-interacting residues. Our trained models were evaluated on an independent dataset of 46 DNA-binding proteins having 965 DNA interacting and 9911 non-interacting residues. All proteins in the independent dataset have less than 30% of sequence similarity with proteins in the training dataset. A wide range of traditional machine learning and deep learning (1D-CNN) techniques-based models have been developed using binary, physicochemical properties and Position-Specific Scoring Matrix (PSSM)/evolutionary profiles. In the case of machine learning technique, eXtreme Gradient Boosting-based model achieved a maximum area under the receiver operating characteristics (AUROC) curve of 0.77 on the independent dataset using PSSM profile. Deep learning-based model achieved the highest AUROC of 0.79 on the independent dataset using a combination of all three profiles. We evaluated the performance of existing methods on the independent dataset and observed that our proposed method outperformed all the existing methods. In order to facilitate scientific community, we developed standalone software and web server, which are accessible from https://webs.iiitd.edu.in/raghava/dbpred.

Copy number variation (CNV) is a class of key biomarkers in many complex traits and diseases. Detecting CNV from sequencing data is a substantial bioinformatics problem and a standard requirement in clinical practice. Although many proposed CNV detection approaches exist, the core statistical model at their foundation is weakened by two critical computational issues: (i) identifying the optimal setting on the sliding window and (ii) correcting for bias and noise. We designed a statistical process model to overcome these limitations by calculating regional read depths via an exponentially weighted moving average strategy. A one-run detection of CNVs of various lengths is then achieved by a dynamic sliding window, whose size is self-adopted according to the weighted averages. We also designed a novel bias/noise reduction model, accompanied by the moving average, which can handle complicated patterns and extend training data. This model, called PEcnv, accurately detects CNVs ranging from kb-scale to chromosome-arm level. The model performance was validated with simulation samples and real samples. Comparative analysis showed that PEcnv outperforms current popular approaches. Notably, PEcnv provided considerable advantages in detecting small CNVs (1 kb-1 Mb) in panel sequencing data. Thus, PEcnv fills the gap left by existing methods focusing on large CNVs. PEcnv may have broad applications in clinical testing where panel sequencing is the dominant strategy. Availability and implementation: Source code is freely available at https://github.com/Sherwin-xjtu/PEcnv.

Coronavirus disease 2019 (COVID-19) has infected hundreds of millions of people and killed millions of them. As an RNA virus, COVID-19 is more susceptible to variation than other viruses. Many problems involved in this epidemic have made biosafety and biosecurity (hereafter collectively referred to as 'biosafety') a popular and timely topic globally. Biosafety research covers a broad and diverse range of topics, and it is important to quickly identify hotspots and trends in biosafety research through big data analysis. However, the data-driven literature on biosafety research discovery is quite scant. We developed a novel topic model based on latent Dirichlet allocation, affinity propagation clustering and the PageRank algorithm (LDAPR) to extract knowledge from biosafety research publications from 2011 to 2020. Then, we conducted hotspot and trend analysis with LDAPR and carried out further studies, including annual hot topic extraction, a 10-year keyword evolution trend analysis, topic map construction, hot region discovery and fine-grained correlation analysis of interdisciplinary research topic trends. These analyses revealed valuable information that can guide epidemic prevention work: (1) the research enthusiasm over a certain infectious disease not only is related to its epidemic characteristics but also is affected by the progress of research on other diseases, and (2) infectious diseases are not only strongly related to their corresponding microorganisms but also potentially related to other specific microorganisms. The detailed experimental results and our code are available at https://github.com/KEAML-JLU/Biosafety-analysis.

The link between tumor genetic variations and immunotherapy benefits has been widely recognized. Recent studies suggested that the key biological pathways activated by accumulated genetic mutations may act as an effective biomarker for predicting the efficacy of immune checkpoint inhibitor (ICI) therapy. Here, we developed a novel individual Pathway Mutation Perturbation (iPMP) method that measures the pathway mutation perturbation level by combining evidence of the cumulative effect of mutated genes with the position of mutated genes in the pathways. In iPMP, somatic mutations on a single sample were first mapped to genes in a single pathway to infer the pathway mutation perturbation score (PMPscore), and then, an integrated PMPscore profile was produced, which can be used in place of the original mutation dataset to identify associations with clinical outcomes. To illustrate the effect of iPMP, we applied it to a melanoma cohort treated with ICIs and identified seven significant perturbation pathways, which jointly constructed a pathway-based signature. With the signature, patients were classified into two subgroups with significant distinctive overall survival and objective response rate to immunotherapy. Moreover, the pathway-based signature was consistently validated in two independent melanoma cohorts. We further applied iPMP to two non-small cell lung cancer cohorts and also obtained good performance. Altogether, the iPMP method could be used to identify the significant mutation perturbation pathways for constructing the pathway-based biomarker to predict the clinical outcomes of immunotherapy. The iPMP method has been implemented as a freely available R-based package (https://CRAN.R-project.org/package=PMAPscore).

Natural products (NPs) and their derivatives are important resources for drug discovery. There are many in silico target prediction methods that have been reported, however, very few of them distinguish NPs from synthetic molecules. Considering the fact that NPs and synthetic molecules are very different in many characteristics, it is necessary to build specific target prediction models of NPs. Therefore, we collected the activity data of NPs and their derivatives from the public databases and constructed four datasets, including the NP dataset, the NPs and its first-class derivatives dataset, the NPs and all its derivatives and the ChEMBL26 compounds dataset. Conditions, including activity thresholds and input features, were explored to access the performance of eight machine learning methods of target prediction of NPs, including support vector machines (SVM), extreme gradient boosting, random forests, K-nearest neighbor, naive Bayes, feedforward neural networks (FNN), convolutional neural networks and recurrent neural networks. As a result, the NPs and all their derivatives datasets were selected to build the best NP-specific models. Furthermore, the consensus models, as well as the voting models, were additionally applied to improve the prediction performance. More evaluations were made on the external validation set and the results demonstrated that (1) the NP-specific model performed better on the target prediction of NPs than the traditional models training on the whole compounds of ChEMBL26. (2) The consensus model of FNN + SVM possessed the best overall performance, and the voting model can significantly improve recall and specificity.

The ability of a compound to permeate across the blood-brain barrier (BBB) is a significant factor for central nervous system drug development. Thus, for speeding up the drug discovery process, it is crucial to perform high-throughput screenings to predict the BBB permeability of the candidate compounds. Although experimental methods are capable of determining BBB permeability, they are still cost-ineffective and time-consuming. To complement the shortcomings of existing methods, we present a deep learning-based multi-model framework model, called Deep-B3, to predict the BBB permeability of candidate compounds. In Deep-B3, the samples are encoded in three kinds of features, namely molecular descriptors and fingerprints, molecular graph and simplified molecular input line entry system (SMILES) text notation. The pre-trained models were built to extract latent features from the molecular graph and SMILES. These features depicted the compounds in terms of tabular data, image and text, respectively. The validation results yielded from the independent dataset demonstrated that the performance of Deep-B3 is superior to that of the state-of-the-art models. Hence, Deep-B3 holds the potential to become a useful tool for drug development. A freely available online web-server for Deep-B3 was established at http://cbcb.cdutcm.edu.cn/deepb3/, and the source code and dataset of Deep-B3 are available at https://github.com/GreatChenLab/Deep-B3.

Noncoding RNAs (ncRNAs) have recently attracted considerable attention due to their key roles in biology. The ncRNA-proteins interaction (NPI) is often explored to reveal some biological activities that ncRNA may affect, such as biological traits, diseases, etc. Traditional experimental methods can accomplish this work but are often labor-intensive and expensive. Machine learning and deep learning methods have achieved great success by exploiting sufficient sequence or structure information. Graph Neural Network (GNN)-based methods consider the topology in ncRNA-protein graphs and perform well on tasks like NPI prediction. Based on GNN, some pairwise constraint methods have been developed to apply on homogeneous networks, but not used for NPI prediction on heterogeneous networks. In this paper, we construct a pairwise constrained NPI predictor based on dual Graph Convolutional Network (GCN) called NPI-DGCN. To our knowledge, our method is the first to train a heterogeneous graph-based model using a pairwise learning strategy. Instead of binary classification, we use a rank layer to calculate the score of an ncRNA-protein pair. Moreover, our model is the first to predict NPIs on the ncRNA-protein bipartite graph rather than the homogeneous graph. We transform the original ncRNA-protein bipartite graph into two homogenous graphs on which to explore second-order implicit relationships. At the same time, we model direct interactions between two homogenous graphs to explore explicit relationships. Experimental results on the four standard datasets indicate that our method achieves competitive performance with other state-of-the-art methods. And the model is available at https://github.com/zhuoninnin1992/NPIPredict.

Advances in single-cell RNA sequencing (scRNA-seq) technologies has provided an unprecedent opportunity for cell-type identification. As clustering is an effective strategy towards cell-type identification, various computational approaches have been proposed for clustering scRNA-seq data. Recently, with the emergence of cellular indexing of transcriptomes and epitopes by sequencing (CITE-seq), the cell surface expression of specific proteins and the RNA expression on the same cell can be captured, which provides more comprehensive information for cell analysis. However, existing single cell clustering algorithms are mainly designed for single-omic data, and have difficulties in handling multi-omics data with diverse characteristics efficiently. In this study, we propose a novel deep embedded multi-omics clustering with collaborative training (DEMOC) model to perform joint clustering on CITE-seq data. Our model can take into account the characteristics of transcriptomic and proteomic data, and make use of the consistent and complementary information provided by different data sources effectively. Experiment results on two real CITE-seq datasets demonstrate that our DEMOC model not only outperforms state-of-the-art single-omic clustering methods, but also achieves better and more stable performance than existing multi-omics clustering methods. We also apply our model on three scRNA-seq datasets to assess the performance of our model in rare cell-type identification, novel cell-subtype detection and cellular heterogeneity analysis. Experiment results illustrate the effectiveness of our model in discovering the underlying patterns of data.

The rapid development of spatial transcriptomics allows the measurement of RNA abundance at a high spatial resolution, making it possible to simultaneously profile gene expression, spatial locations of cells or spots, and the corresponding hematoxylin and eosin-stained histology images. It turns promising to predict gene expression from histology images that are relatively easy and cheap to obtain. For this purpose, several methods are devised, but they have not fully captured the internal relations of the 2D vision features or spatial dependency between spots. Here, we developed Hist2ST, a deep learning-based model to predict RNA-seq expression from histology images. Around each sequenced spot, the corresponding histology image is cropped into an image patch and fed into a convolutional module to extract 2D vision features. Meanwhile, the spatial relations with the whole image and neighbored patches are captured through Transformer and graph neural network modules, respectively. These learned features are then used to predict the gene expression by following the zero-inflated negative binomial distribution. To alleviate the impact by the small spatial transcriptomics data, a self-distillation mechanism is employed for efficient learning of the model. By comprehensive tests on cancer and normal datasets, Hist2ST was shown to outperform existing methods in terms of both gene expression prediction and spatial region identification. Further pathway analyses indicated that our model could reserve biological information. Thus, Hist2ST enables generating spatial transcriptomics data from histology images for elucidating molecular signatures of tissues.

Adoption of recently developed methods from machine learning has given rise to creation of drug-discovery knowledge graphs (KGs) that utilize the interconnected nature of the domain. Graph-based modelling of the data, combined with KG embedding (KGE) methods, are promising as they provide a more intuitive representation and are suitable for inference tasks such as predicting missing links. One common application is to produce ranked lists of genes for a given disease, where the rank is based on the perceived likelihood of association between the gene and the disease. It is thus critical that these predictions are not only pertinent but also biologically meaningful. However, KGs can be biased either directly due to the underlying data sources that are integrated or due to modelling choices in the construction of the graph, one consequence of which is that certain entities can get topologically overrepresented. We demonstrate the effect of these inherent structural imbalances, resulting in densely connected entities being highly ranked no matter the context. We provide support for this observation across different datasets, models as well as predictive tasks. Further, we present various graph perturbation experiments which yield more support to the observation that KGE models can be more influenced by the frequency of entities rather than any biological information encoded within the relations. Our results highlight the importance of data modelling choices, and emphasizes the need for practitioners to be mindful of these issues when interpreting model outputs and during KG composition.

Structural variations (SVs) play important roles in human genetic diversity; deletions and insertions are two common types of SVs that have been proven to be associated with genetic diseases. Hence, accurately detecting and genotyping SVs is significant for disease research. Despite the fact that long-read sequencing technologies have improved the field of SV detection and genotyping, there are still some challenges that prevent satisfactory results from being obtained. In this paper, we propose MAMnet, a fast and scalable SV detection and genotyping method based on long reads and a combination of convolutional neural network and long short-term network. MAMnet uses a deep neural network to implement sensitive SV detection with a novel prediction strategy. On real long-read sequencing datasets, we demonstrate that MAMnet outperforms Sniffles, SVIM, cuteSV and PBSV in terms of their F1 scores while achieving better scaling performance. The source code is available from https://github.com/micahvista/MAMnet.

Single-cell RNA-sequencing (scRNA-seq) enables researchers to quantify transcriptomes of thousands of cells simultaneously and study transcriptomic changes between cells. scRNA-seq datasets increasingly include multisubject, multicondition experiments to investigate cell-type-specific differential states (DS) between conditions. This can be performed by first identifying the cell types in all the subjects and then by performing a DS analysis between the conditions within each cell type. Naïve single-cell DS analysis methods that treat cells statistically independent are subject to false positives in the presence of variation between biological replicates, an issue known as the pseudoreplicate bias. While several methods have already been introduced to carry out the statistical testing in multisubject scRNA-seq analysis, comparisons that include all these methods are currently lacking. Here, we performed a comprehensive comparison of 18 methods for the identification of DS changes between conditions from multisubject scRNA-seq data. Our results suggest that the pseudobulk methods performed generally best. Both pseudobulks and mixed models that model the subjects as a random effect were superior compared with the naïve single-cell methods that do not model the subjects in any way. While the naïve models achieved higher sensitivity than the pseudobulk methods and the mixed models, they were subject to a high number of false positives. In addition, accounting for subjects through latent variable modeling did not improve the performance of the naïve methods.

A large amount of clinical evidence began to mount, showing that circular ribonucleic acids (RNAs; circRNAs) perform a very important function in complex diseases by participating in transcription and translation regulation of microRNA (miRNA) target genes. However, with strict high-throughput techniques based on traditional biological experiments and the conditions and environment, the association between circRNA and miRNA can be discovered to be labor-intensive, expensive, time-consuming, and inefficient. In this paper, we proposed a novel computational model based on Word2vec, Structural Deep Network Embedding (SDNE), Convolutional Neural Network and Deep Neural Network, which predicts the potential circRNA-miRNA associations, called Word2vec, SDNE, Convolutional Neural Network and Deep Neural Network (WSCD). Specifically, the WSCD model extracts attribute feature and behaviour feature by word embedding and graph embedding algorithm, respectively, and ultimately feed them into a feature fusion model constructed by combining Convolutional Neural Network and Deep Neural Network to deduce potential circRNA-miRNA interactions. The proposed method is proved on dataset and obtained a prediction accuracy and an area under the receiver operating characteristic curve of 81.61% and 0.8898, respectively, which is shown to have much higher accuracy than the state-of-the-art models and classifier models in prediction. In addition, 23 miRNA-related circular RNAs (circRNAs) from the top 30 were confirmed in relevant experiences. In these works, all results represent that WSCD would be a helpful supplementary reliable method for predicting potential miRNA-circRNA associations compared to wet laboratory experiments.

The coronavirus disease 2019 pandemic has alerted people of the threat caused by viruses. Vaccine is the most effective way to prevent the disease from spreading. The interaction between antibodies and antigens will clear the infectious organisms from the host. Identifying B-cell epitopes is critical in vaccine design, development of disease diagnostics and antibody production. However, traditional experimental methods to determine epitopes are time-consuming and expensive, and the predictive performance using the existing in silico methods is not satisfactory. This paper develops a general framework to predict variable-length linear B-cell epitopes specific for human-adapted viruses with machine learning approaches based on Protvec representation of peptides and physicochemical properties of amino acids. QR decomposition is incorporated during the embedding process that enables our models to handle variable-length sequences. Experimental results on large immune epitope datasets validate that our proposed model's performance is superior to the state-of-the-art methods in terms of AUROC (0.827) and AUPR (0.831) on the testing set. Moreover, sequence analysis also provides the results of the viral category for the corresponding predicted epitopes with high precision. Therefore, this framework is shown to reliably identify linear B-cell epitopes of human-adapted viruses given protein sequences and could provide assistance for potential future pandemics and epidemics.

MOTIVATION: Long noncoding RNAs (lncRNAs) play an important role in the occurrence and development of diseases. Predicting disease-related lncRNAs can help to understand the pathogenesis of diseases deeply. The existing methods mainly rely on multi-source data related to lncRNAs and diseases when predicting the associations between lncRNAs and diseases. There are interdependencies among node attributes in a heterogeneous graph composed of all lncRNAs, diseases and micro RNAs. The meta-paths composed of various connections between them also contain rich semantic information. However, the existing methods neglect to integrate attribute information of intermediate nodes in meta-paths. RESULTS: We propose a novel association prediction model, GSMV, to learn and deeply integrate the global dependencies, semantic information of meta-paths and node-pair multi-view features related to lncRNAs and diseases. We firstly formulate the global representations of the lncRNA and disease nodes by establishing a self-attention mechanism to capture and learn the global dependencies among node attributes. Second, starting from the lncRNA and disease nodes, respectively, multiple meta-pathways are established to reveal different semantic information. Considering that each meta-path contains specific semantics and has multiple meta-path instances which have different contributions to revealing meta-path semantics, we design a graph neural network based module which consists of a meta-path instance encoding strategy and two novel attention mechanisms. The proposed meta-path instance encoding strategy is used to learn the contextual connections between nodes within a meta-path instance. One of the two new attention mechanisms is at the meta-path instance level, which learns rich and informative meta-path instances. The other attention mechanism integrates various semantic information from multiple meta-paths to learn the semantic representation of lncRNA and disease nodes. Finally, a dilated convolution-based learning module with adjustable receptive fields is proposed to learn multi-view features of lncRNA-disease node pairs. The experimental results prove that our method outperforms seven state-of-the-art comparing methods for lncRNA-disease association prediction. Ablation experiments demonstrate the contributions of the proposed global representation learning, semantic information learning, pairwise multi-view feature learning and the meta-path instance encoding strategy. Case studies on three cancers further demonstrate our method's ability to discover potential disease-related lncRNA candidates. CONTACT: zhang@hlju.edu.cn or peiliangwu@ysu.edu.cn. SUPPLEMENTARY INFORMATION: Supplementary data are available at Briefings in Bioinformatics online.

With the development of high-throughput technologies, the accumulation of large amounts of multidimensional genomic data provides an excellent opportunity to study the multilevel biological regulatory relationships in cancer. Based on the hypothesis of competitive endogenous ribonucleic acid (RNA) (ceRNA) network, lncRNAs can eliminate the inhibition of microRNAs (miRNAs) on their target genes by binding to intracellular miRNA sites so as to improve the expression level of these target genes. However, previous studies on cancer expression mechanism are mostly based on individual or two-dimensional data, and lack of integration and analysis of various RNA-seq data, making it difficult to verify the complex biological relationships involved. To explore RNA expression patterns and potential molecular mechanisms of cancer, a network-regularized sparse orthogonal-regularized joint non-negative matrix factorization (NSOJNMF) algorithm is proposed, which combines the interaction relations among RNA-seq data in the way of network regularization and effectively prevents multicollinearity through sparse constraints and orthogonal regularization constraints to generate good modular sparse solutions. NSOJNMF algorithm is performed on the datasets of liver cancer and colon cancer, then ceRNA co-modules of them are recognized. The enrichment analysis of these modules shows that >90% of them are closely related to the occurrence and development of cancer. In addition, the ceRNA networks constructed by the ceRNA co-modules not only accurately mine the known correlations of the three RNA molecules but also further discover their potential biological associations, which may contribute to the exploration of the competitive relationships among multiple RNAs and the molecular mechanisms affecting tumor development.

Cell types (subpopulations) serve as bio-markers for the diagnosis and therapy of complex diseases, and single-cell RNA-sequencing (scRNA-seq) measures expression of genes at cell level, paving the way for the identification of cell types. Although great efforts have been devoted to this issue, it remains challenging to identify rare cell types in scRNA-seq data because of the few-shot problem, lack of interpretability and separation of generating samples and clustering of cells. To attack these issues, a novel deep generative model for leveraging the small samples of cells (aka scLDS2) is proposed by precisely estimating the distribution of different cells, which discriminate the rare and non-rare cell types with adversarial learning. Specifically, to enhance interpretability of samples, scLDS2 generates the sparse faked samples of cells with $\ell _1$-norm, where the relations among cells are learned, facilitating the identification of cell types. Furthermore, scLDS2 directly obtains cell types from the generated samples by learning the block structure such that cells belonging to the same types are similar to each other with the nuclear-norm. scLDS2 joins the generation of samples, classification of the generated and truth samples for cells and feature extraction into a unified generative framework, which transforms the rare cell types detection problem into a classification problem, paving the way for the identification of cell types with joint learning. The experimental results on 20 datasets demonstrate that scLDS2 significantly outperforms 17 state-of-the-art methods in terms of various measurements with 25.12% improvement in adjusted rand index on average, providing an effective strategy for scRNA-seq data with rare cell types. (The software is coded using python, and is freely available for academic https://github.com/xkmaxidian/scLDS2).

Prokaryotic viruses, which infect bacteria and archaea, are key players in microbial communities. Predicting the hosts of prokaryotic viruses helps decipher the dynamic relationship between microbes. Experimental methods for host prediction cannot keep pace with the fast accumulation of sequenced phages. Thus, there is a need for computational host prediction. Despite some promising results, computational host prediction remains a challenge because of the limited known interactions and the sheer amount of sequenced phages by high-throughput sequencing technologies. The state-of-the-art methods can only achieve 43% accuracy at the species level. In this work, we formulate host prediction as link prediction in a knowledge graph that integrates multiple protein and DNA-based sequence features. Our implementation named CHERRY can be applied to predict hosts for newly discovered viruses and to identify viruses infecting targeted bacteria. We demonstrated the utility of CHERRY for both applications and compared its performance with 11 popular host prediction methods. To our best knowledge, CHERRY has the highest accuracy in identifying virus-prokaryote interactions. It outperforms all the existing methods at the species level with an accuracy increase of 37%. In addition, CHERRY's performance on short contigs is more stable than other tools.

MOTIVATION: Accurate identification of proteins interacted with drugs helps reduce the time and cost of drug development. Most of previous methods focused on integrating multisource data about drugs and proteins for predicting drug-target interactions (DTIs). There are both similarity connection and interaction connection between two drugs, and these connections reflect their relationships from different perspectives. Similarly, two proteins have various connections from multiple perspectives. However, most of previous methods failed to deeply integrate these connections. In addition, multiple drug-protein heterogeneous networks can be constructed based on multiple kinds of connections. The diverse topological structures of these networks are still not exploited completely. RESULTS: We propose a novel model to extract and integrate multi-type neighbor topology information, diverse similarities and interactions related to drugs and proteins. Firstly, multiple drug-protein heterogeneous networks are constructed according to multiple kinds of connections among drugs and those among proteins. The multi-type neighbor node sequences of a drug node (or a protein node) are formed by random walks on each network and they reflect the hidden neighbor topological structure of the node. Secondly, a module based on graph neural network (GNN) is proposed to learn the multi-type neighbor topologies of each node. We propose attention mechanisms at neighbor node level and at neighbor type level to learn more informative neighbor nodes and neighbor types. A network-level attention is also designed to enhance the context dependency among multiple neighbor topologies of a pair of drug and protein nodes. Finally, the attribute embedding of the drug-protein pair is formulated by a proposed embedding strategy, and the embedding covers the similarities and interactions about the pair. A module based on three-dimensional convolutional neural networks (CNN) is constructed to deeply integrate pairwise attributes. Extensive experiments have been performed and the results indicate GCDTI outperforms several state-of-the-art prediction methods. The recall rate estimation over the top-ranked candidates and case studies on 5 drugs further demonstrate GCDTI's ability in discovering potential drug-protein interactions.

Antimicrobial peptides (AMPs) are a heterogeneous group of short polypeptides that target not only microorganisms but also viruses and cancer cells. Due to their lower selection for resistance compared with traditional antibiotics, AMPs have been attracting the ever-growing attention from researchers, including bioinformaticians. Machine learning represents the most cost-effective method for novel AMP discovery and consequently many computational tools for AMP prediction have been recently developed. In this article, we investigate the impact of negative data sampling on model performance and benchmarking. We generated 660 predictive models using 12 machine learning architectures, a single positive data set and 11 negative data sampling methods; the architectures and methods were defined on the basis of published AMP prediction software. Our results clearly indicate that similar training and benchmark data set, i.e. produced by the same or a similar negative data sampling method, positively affect model performance. Consequently, all the benchmark analyses that have been performed for AMP prediction models are significantly biased and, moreover, we do not know which model is the most accurate. To provide researchers with reliable information about the performance of AMP predictors, we also created a web server AMPBenchmark for fair model benchmarking. AMPBenchmark is available at http://BioGenies.info/AMPBenchmark.

Biological pathways are a broadly used formalism for representing and interpreting the cascade of biochemical reactions underlying cellular and biological mechanisms. Pathway representation provides an ontological link among biomolecules such as RNA, DNA, small molecules, proteins, protein complexes, hormones and genes. Frequently, pathway annotations are used to identify mechanisms linked to genes within affected biological contexts. This important role and the simplicity and elegance in representing complex interactions led to an explosion of pathway representations and databases. Unfortunately, the lack of overlap across databases results in inconsistent enrichment analysis results, unless databases are integrated. However, due to absence of consensus, guidelines or gold standards in pathway definition and representation, integration of data across pathway databases is not straightforward. Despite multiple attempts to provide consolidated pathways, highly related, redundant, poorly overlapping or ambiguous pathways continue to render pathways analysis inconsistent and hard to interpret. Ontology-based integration will promote unbiased, comprehensive yet streamlined analysis of experiments, and will reduce the number of enriched pathways when performing pathway enrichment analysis. Moreover, appropriate and consolidated pathways provide better training data for pathway prediction algorithms. In this manuscript, we describe the current methods for pathway consolidation, their strengths and pitfalls, and highlight directions for future improvements to this research area.

CRISPR/Cas9 system is widely used in a broad range of gene-editing applications. While this editing technique is quite accurate in the target region, there may be many unplanned off-target sites (OTSs). Consequently, a plethora of computational methods have been developed to predict off-target cleavage sites given a guide RNA and a reference genome. However, these methods are based on small-scale datasets (only tens to hundreds of OTSs) produced by experimental techniques to detect OTSs with a low signal-to-noise ratio. Recently, CHANGE-seq, a new in vitro experimental technique to detect OTSs, was used to produce a dataset of unprecedented scale and quality (>200 000 OTS over 110 guide RNAs). In addition, the same study included in cellula GUIDE-seq experiments for 58 of the guide RNAs. Here, we fill the gap in previous computational methods by utilizing these data to systematically evaluate data processing and formulation of the CRISPR OTSs prediction problem. Our evaluations show that data transformation as a pre-processing phase is critical prior to model training. Moreover, we demonstrate the improvement gained by adding potential inactive OTSs to the training datasets. Furthermore, our results point to the importance of adding the number of mismatches between guide RNAs and their OTSs as a feature. Finally, we present predictive off-target in cellula models based on both in vitro and in cellula data and compare them to state-of-the-art methods in predicting true OTSs. Our conclusions will be instrumental in any future development of an off-target predictor based on high-throughput datasets.

Finding personalized biomarkers for disease prediction of patients with cancer remains a massive challenge in precision medicine. Most methods focus on one subnetwork or module as a network biomarker; however, this ignores the early warning capabilities of other modules with different configurations of biomarkers (i.e. multi-modal personalized biomarkers). Identifying such modules would not only predict disease but also provide effective therapeutic drug target information for individual patients. To solve this problem, we developed a novel model (denoted multi-modal personalized dynamic network biomarkers (MMPDNB)) based on a multi-modal optimization mechanism and personalized dynamic network biomarker (PDNB) theory, which can provide multiple modules of personalized biomarkers and unveil their multi-modal properties. Using the genomics data of patients with breast or lung cancer from The Cancer Genome Atlas database, we validated the effectiveness of the MMPDNB model. The experimental results showed that compared with other advanced methods, MMPDNB can more effectively predict the critical state with the highest early warning signal score during cancer development. Furthermore, MMPDNB more significantly identified PDNBs containing driver and biomarker genes specific to cancer tissues. More importantly, we validated the biological significance of multi-modal PDNBs, which could provide effective drug targets of individual patients as well as markers for predicting early warning signals of the critical disease state. In conclusion, multi-modal optimization is an effective method to identify PDNBs and offers a new perspective for understanding tumor heterogeneity in cancer precision medicine.

Increasing biomedical evidence has proved that the dysregulation of miRNAs is associated with human complex diseases. Identification of disease-related miRNAs is of great importance for disease prevention, diagnosis and remedy. To reduce the time and cost of biomedical experiments, there is a strong incentive to develop efficient computational methods to infer potential miRNA-disease associations. Although many computational approaches have been proposed to address this issue, the prediction accuracy needs to be further improved. In this study, we present a computational framework MKGAT to predict possible associations between miRNAs and diseases through graph attention networks (GATs) using dual Laplacian regularized least squares. We use GATs to learn embeddings of miRNAs and diseases on each layer from initial input features of known miRNA-disease associations, intra-miRNA similarities and intra-disease similarities. We then calculate kernel matrices of miRNAs and diseases based on Gaussian interaction profile (GIP) with the learned embeddings. We further fuse the kernel matrices of each layer and initial similarities with attention mechanism. Dual Laplacian regularized least squares are finally applied for new miRNA-disease association predictions with the fused miRNA and disease kernels. Compared with six state-of-the-art methods by 5-fold cross-validations, our method MKGAT receives the highest AUROC value of 0.9627 and AUPR value of 0.7372. We use MKGAT to predict related miRNAs for three cancers and discover that all the top 50 predicted results in the three diseases are confirmed by existing databases. The excellent performance indicates that MKGAT would be a useful computational tool for revealing disease-related miRNAs.

Erratum in Brief Bioinform. 2022 Nov 19;23(6):

Liquid biopsy has shown promise for cancer diagnosis due to its minimally invasive nature and the potential for novel biomarker discovery. However, the low concentration of relevant blood-based biosources and the heterogeneity of samples (i.e. the variability of relative abundance of molecules identified), pose major challenges to biomarker discovery. Moreover, the number of molecular measurements or features (e.g. transcript read counts) per sample could be in the order of several thousand, whereas the number of samples is often substantially lower, leading to the curse of dimensionality. These challenges, among others, elucidate the importance of a robust biomarker panel identification or feature extraction step wherein relevant molecular measurements are identified prior to classification for cancer detection. In this work, we performed a benchmarking study on 12 feature extraction methods using transcriptomic profiles derived from different blood-based biosources. The methods were assessed both in terms of their predictive performance and the robustness of the biomarker panels in diagnosing cancer or stratifying cancer subtypes. While performing the comparison, the feature extraction methods are categorized into feature subset selection methods and transformation methods. A transformation feature extraction method, namely partial least square discriminant analysis, was found to perform consistently superior in terms of classification performance. As part of the benchmarking study, a generic pipeline has been created and made available as an R package to ensure reproducibility of the results and allow for easy extension of this study to other datasets (https://github.com/VafaeeLab/bloodbased-pancancer-diagnosis).

RNA 5-hydroxymethylcytosine (5hmC) is a kind of RNA modification, which is related to the life activities of many organisms. Studying its distribution is very important to reveal its biological function. Previously, high-throughput sequencing was used to identify 5hmC, but it is expensive and inefficient. Therefore, machine learning is used to identify 5hmC sites. Here, we design a model called R5hmCFDV, which is mainly divided into feature representation, feature fusion and classification. (i) Pseudo dinucleotide composition, dinucleotide binary profile and frequency, natural vector and physicochemical property are used to extract features from four aspects: nucleotide composition, coding, natural language and physical and chemical properties. (ii) To strengthen the relevance of features, we construct a novel feature fusion method. Firstly, the attention mechanism is employed to process four single features, stitch them together and feed them to the convolution layer. After that, the output data are processed by BiGRU and BiLSTM, respectively. Finally, the features of these two parts are fused by the multiply function. (iii) We design the deep voting algorithm for classification by imitating the soft voting mechanism in the Python package. The base classifiers contain deep neural network (DNN), convolutional neural network (CNN) and improved gated recurrent unit (GRU). And then using the principle of soft voting, the corresponding weights are assigned to the predicted probabilities of the three classifiers. The predicted probability values are multiplied by the corresponding weights and then summed to obtain the final prediction results. We use 10-fold cross-validation to evaluate the model, and the evaluation indicators are significantly improved. The prediction accuracy of the two datasets is as high as 95.41% and 93.50%, respectively. It demonstrates the stronger competitiveness and generalization performance of our model. In addition, all datasets and source codes can be found at https://github.com/HongyanShi026/R5hmCFDV.

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.

Circular RNA (circRNA) is closely involved in physiological and pathological processes of many diseases. Discovering the associations between circRNAs and diseases is of great significance. Due to the high-cost to verify the circRNA-disease associations by wet-lab experiments, computational approaches for predicting the associations become a promising research direction. In this paper, we propose a method, MDGF-MCEC, based on multi-view dual attention graph convolution network (GCN) with cooperative ensemble learning to predict circRNA-disease associations. First, MDGF-MCEC constructs two disease relation graphs and two circRNA relation graphs based on different similarities. Then, the relation graphs are fed into a multi-view GCN for representation learning. In order to learn high discriminative features, a dual-attention mechanism is introduced to adjust the contribution weights, at both channel level and spatial level, of different features. Based on the learned embedding features of diseases and circRNAs, nine different feature combinations between diseases and circRNAs are treated as new multi-view data. Finally, we construct a multi-view cooperative ensemble classifier to predict the associations between circRNAs and diseases. Experiments conducted on the CircR2Disease database demonstrate that the proposed MDGF-MCEC model achieves a high area under curve of 0.9744 and outperforms the state-of-the-art methods. Promising results are also obtained from experiments on the circ2Disease and circRNADisease databases. Furthermore, the predicted associated circRNAs for hepatocellular carcinoma and gastric cancer are supported by the literature. The code and dataset of this study are available at https://github.com/ABard0/MDGF-MCEC.

The X-ray diffraction (XRD) technique based on crystallography is the main experimental method to analyze the three-dimensional structure of proteins. The production process of protein crystals on which the XRD technique relies has undergone multiple experimental steps, which requires a lot of manpower and material resources. In addition, studies have shown that not all proteins can form crystals under experimental conditions, and the success rate of the final crystallization of proteins is only <10%. Although some protein crystallization predictors have been developed, not many tools capable of predicting multi-stage protein crystallization propensity are available and the accuracy of these tools is not satisfactory. In this paper, we propose a novel deep learning framework, named SADeepcry, for predicting protein crystallization propensity. The framework can be used to estimate the three steps (protein material production, purification and crystallization) in protein crystallization experiments and the success rate of the final protein crystallization. SADeepcry uses the optimized self-attention and auto-encoder modules to extract sequence, structure and physicochemical features from the proteins. Compared with other state-of-the-art protein crystallization propensity prediction models, SADeepcry can obtain more complex global spatial long-distance dependence of protein sequence information. Our computational results show that SADeepcry has increased Matthews correlation coefficient and area under the curve, by 100.3% and 13.4%, respectively, over the DCFCrystal method on the benchmark dataset. The codes of SADeepcry are available at https://github.com/zhc940702/SADeepcry.

Nuclear receptors (NRs) are ligand-activated transcription factors, which constitute one of the most important targets for drug discovery. Current computational strategies mainly focus on a single target, and the transfer of learned knowledge among NRs was not considered yet. Herein we proposed a novel computational framework named NR-Profiler for prediction of potential NR modulators with high affinity and specificity. First, we built a comprehensive NR data set including 42 684 interactions to connect 42 NRs and 31 033 compounds. Then, we used multi-task deep neural network and multi-task graph convolutional neural network architectures to construct multi-task multi-classification models. To improve the predictive capability and robustness, we built a consensus model with an area under the receiver operating characteristic curve (AUC) = 0.883. Compared with conventional machine learning and structure-based approaches, the consensus model showed better performance in external validation. Using this consensus model, we demonstrated the practical value of NR-Profiler in virtual screening for NRs. In addition, we designed a selectivity score to quantitatively measure the specificity of NR modulators. Finally, we developed a freely available standalone software for users to make profiling predictions for their compounds of interest. In summary, our NR-Profiler provides a useful tool for NR-profiling prediction and is expected to facilitate NR-based drug discovery.

Posttranslational modification of lysine residues, K-PTM, is one of the most popular PTMs. Some lysine residues in proteins can be continuously or cascaded covalently modified, such as acetylation, crotonylation, methylation and succinylation modification. The covalent modification of lysine residues may have some special functions in basic research and drug development. Although many computational methods have been developed to predict lysine PTMs, up to now, the K-PTM prediction methods have been modeled and learned a single class of K-PTM modification. In view of this, this study aims to fill this gap by building a multi-label computational model that can be directly used to predict multiple K-PTMs in proteins. In this study, a multi-label prediction model, MLysPRED, is proposed to identify multiple lysine sites using features generated from human protein sequences. In MLysPRED, three kinds of multi-label sequence encoding algorithms (MLDBPB, MLPSDAAP, MLPSTAAP) are proposed and combined with three encoding strategies (CHHAA, DR and Kmer) to convert preprocessed lysine sequences into effective numerical features. A multidimensional normal distribution oversampling technique and graph-based multi-view clustering under-sampling algorithm were first proposed and incorporated to reduce the proportion of the original training samples, and multi-label nearest neighbor algorithm is used for classification. It is observed that MLysPRED achieved an Aiming of 92.21%, Coverage of 94.98%, Accuracy of 89.63%, Absolute-True of 81.46% and Absolute-False of 0.0682 on the independent datasets. Additionally, comparison of results with five existing predictors also indicated that MLysPRED is very promising and encouraging to predict multiple K-PTMs in proteins. For the convenience of the experimental scientists, 'MLysPRED' has been deployed as a user-friendly web-server at http://47.100.136.41:8181.

Extracellular vesicles (EVs) carrying various small non-coding RNAs (sncRNAs) play a vital roles in cell communication and diseases. Correct quantification of multiple sncRNA biotypes simultaneously in EVs is a challenge due to the short reads (<30 bp) could be mapped to multiple sncRNA types. To address this question, we developed an optimized reads assignment algorithm (ORAA) to dynamically map multi-mapping reads to the sncRNA type with a higher proportion. We integrated ORAA with reads processing steps into EVAtool Python-package (http://bioinfo.life.hust.edu.cn/EVAtool) to quantify sncRNAs, especially for sncRNA-seq from EV samples. EVAtool allows users to specify interested sncRNA types in advanced mode or use default seven sncRNAs (microRNA, small nucleolar RNA, PIWI-interacting RNAs, small nuclear RNA, ribosomal RNA, transfer RNA and Y RNA). To prove the utilities of EVAtool, we quantified the sncRNA expression profiles for 200 samples from cognitive decline and multiple sclerosis. We found that more than 20% of short reads on average were mapped to multiple sncRNA biotypes in multiple sclerosis. In cognitive decline, the proportion of Y RNA is significantly higher than other sncRNA types. EVAtool is a flexible and extensible tool that would benefit to mine potential biomarkers and functional molecules in EVs.

As recently demonstrated by the COVID-19 pandemic, large-scale pathogen genomic data are crucial to characterize transmission patterns of human infectious diseases. Yet, current methods to process raw sequence data into analysis-ready variants remain slow to scale, hampering rapid surveillance efforts and epidemiological investigations for disease control. Here, we introduce an accelerated, scalable, reproducible, and cost-effective framework for pathogen genomic variant identification and present an evaluation of its performance and accuracy across benchmark datasets of Plasmodium falciparum malaria genomes. We demonstrate superior performance of the GPU framework relative to standard pipelines with mean execution time and computational costs reduced by 27× and 4.6×, respectively, while delivering 99.9% accuracy at enhanced reproducibility.

In recent years, a number of computational approaches have been proposed to effectively integrate multiple heterogeneous biological networks, and have shown impressive performance for inferring gene function. However, the previous methods do not fully represent the critical neighborhood relationship between genes during the feature learning process. Furthermore, it is difficult to accurately estimate the contributions of different views for multi-view integration. In this paper, we propose MGEGFP, a multi-view graph embedding method based on adaptive estimation with Graph Convolutional Network (GCN), to learn high-quality gene representations among multiple interaction networks for function prediction. First, we design a dual-channel GCN encoder to disentangle the view-specific information and the consensus pattern across diverse networks. By the aid of disentangled representations, we develop a multi-gate module to adaptively estimate the contributions of different views during each reconstruction process and make full use of the multiplexity advantages, where a diversity preservation constraint is designed to prevent the over-fitting problem. To validate the effectiveness of our model, we conduct experiments on networks from the STRING database for both yeast and human datasets, and compare the performance with seven state-of-the-art methods in five evaluation metrics. Moreover, the ablation study manifests the important contribution of the designed dual-channel encoder, multi-gate module and the diversity preservation constraint in MGEGFP. The experimental results confirm the superiority of our proposed method and suggest that MGEGFP can be a useful tool for gene function prediction.

MOTIVATION: CircularRNA (circRNA) is a class of noncoding RNA with high conservation and stability, which is considered as an important disease biomarker and drug target. Accumulating pieces of evidence have indicated that circRNA plays a crucial role in the pathogenesis and progression of many complex diseases. As the biological experiments are time-consuming and labor-intensive, developing an accurate computational prediction method has become indispensable to identify disease-related circRNAs. RESULTS: We presented a hybrid graph representation learning framework, named GraphCDA, for predicting the potential circRNA-disease associations. Firstly, the circRNA-circRNA similarity network and disease-disease similarity network were constructed to characterize the relationships of circRNAs and diseases, respectively. Secondly, a hybrid graph embedding model combining Graph Convolutional Networks and Graph Attention Networks was introduced to learn the feature representations of circRNAs and diseases simultaneously. Finally, the learned representations were concatenated and employed to build the prediction model for identifying the circRNA-disease associations. A series of experimental results demonstrated that GraphCDA outperformed other state-of-the-art methods on several public databases. Moreover, GraphCDA could achieve good performance when only using a small number of known circRNA-disease associations as the training set. Besides, case studies conducted on several human diseases further confirmed the prediction capability of GraphCDA for predicting potential disease-related circRNAs. In conclusion, extensive experimental results indicated that GraphCDA could serve as a reliable tool for exploring the regulatory role of circRNAs in complex diseases.

Neural network (NN)-based protein modeling methods have improved significantly in recent years. Although the overall accuracy of the two non-homology-based modeling methods, AlphaFold and RoseTTAFold, is outstanding, their performance for specific protein families has remained unexamined. G-protein-coupled receptor (GPCR) proteins are particularly interesting since they are involved in numerous pathways. This work directly compares the performance of these novel deep learning-based protein modeling methods for GPCRs with the most widely used template-based software-Modeller. We collected the experimentally determined structures of 73 GPCRs from the Protein Data Bank. The official AlphaFold repository and RoseTTAFold web service were used with default settings to predict five structures of each protein sequence. The predicted models were then aligned with the experimentally solved structures and evaluated by the root-mean-square deviation (RMSD) metric. If only looking at each program's top-scored structure, Modeller had the smallest average modeling RMSD of 2.17 Å, which is better than AlphaFold's 5.53 Å and RoseTTAFold's 6.28 Å, probably since Modeller already included many known structures as templates. However, the NN-based methods (AlphaFold and RoseTTAFold) outperformed Modeller in 21 and 15 out of the 73 cases with the top-scored model, respectively, where no good templates were available for Modeller. The larger RMSD values generated by the NN-based methods were primarily due to the differences in loop prediction compared to the crystal structures.

Multi-drug combinations for the treatment of complex diseases are gradually becoming an important treatment, and this type of treatment can take advantage of the synergistic effects among drugs. However, drug-drug interactions (DDIs) are not just all beneficial. Accurate and rapid identifications of the DDIs are essential to enhance the effectiveness of combination therapy and avoid unintended side effects. Traditional DDIs prediction methods use only drug sequence information or drug graph information, which ignores information about the position of atoms and edges in the spatial structure. In this paper, we propose Molormer, a method based on a lightweight attention mechanism for DDIs prediction. Molormer takes the two-dimension (2D) structures of drugs as input and encodes the molecular graph with spatial information. Besides, Molormer uses lightweight-based attention mechanism and self-attention distilling to process spatially the encoded molecular graph, which not only retains the multi-headed attention mechanism but also reduces the computational and storage costs. Finally, we use the Siamese network architecture to serve as the architecture of Molormer, which can make full use of the limited data to train the model for better performance and also limit the differences to some extent between networks dealing with drug features. Experiments show that our proposed method outperforms state-of-the-art methods in Accuracy, Precision, Recall and F1 on multi-label DDIs dataset. In the case study section, we used Molormer to make predictions of new interactions for the drugs Aliskiren, Selexipag and Vorapaxar and validated parts of the predictions. Code and models are available at https://github.com/IsXudongZhang/Molormer.

In the development of targeted drugs, anticancer peptides (ACPs) have attracted great attention because of their high selectivity, low toxicity and minimal non-specificity. In this work, we report a framework of ACPs generation, which combines Wasserstein autoencoder (WAE) generative model and Particle Swarm Optimization (PSO) forward search algorithm guided by attribute predictive model to generate ACPs with desired properties. It is well known that generative models based on Variational AutoEncoder (VAE) and Generative Adversarial Networks (GAN) are difficult to be used for de novo design due to the problems of posterior collapse and difficult convergence of training. Our WAE-based generative model trains more successfully (lower perplexity and reconstruction loss) than both VAE and GAN-based generative models, and the semantic connections in the latent space of WAE accelerate the process of forward controlled generation of PSO, while VAE fails to capture this feature. Finally, we validated our pipeline on breast cancer targets (HIF-1) and lung cancer targets (VEGR, ErbB2), respectively. By peptide-protein docking, we found candidate compounds with the same binding sites as the peptides carried in the crystal structure but with higher binding affinity and novel structures, which may be potent antagonists that interfere with these target-mediated signaling.

Recently, N6-methylation (m6A) has recently become a hot topic due to its key role in disease pathogenesis. Identifying disease-related m6A sites aids in the understanding of the molecular mechanisms and biosynthetic pathways underlying m6A-mediated diseases. Existing methods treat it primarily as a binary classification issue, focusing solely on whether an m6A-disease association exists or not. Although they achieved good results, they all shared one common flaw: they ignored the post-transcriptional regulation events during disease pathogenesis, which makes biological interpretation unsatisfactory. Thus, accurate and explainable computational models are required to unveil the post-transcriptional regulation mechanisms of disease pathogenesis mediated by m6A modification, rather than simply inferring whether the m6A sites cause disease or not. Emerging laboratory experiments have revealed the interactions between m6A and other post-transcriptional regulation events, such as circular RNA (circRNA) targeting, microRNA (miRNA) targeting, RNA-binding protein binding and alternative splicing events, etc., present a diverse landscape during tumorigenesis. Based on these findings, we proposed a low-rank tensor completion-based method to infer disease-related m6A sites from a biological standpoint, which can further aid in specifying the post-transcriptional machinery of disease pathogenesis. It is so exciting that our biological analysis results show that Coronavirus disease 2019 may play a role in an m6A- and miRNA-dependent manner in inducing non-small cell lung cancer.

Linear B-cell epitopes have a prominent role in the development of peptide-based vaccines and disease diagnosis. High variability in the length of these epitopes is a major reason for low accuracy in their prediction. Most of the B-cell epitope prediction methods considered fixed length of epitope sequences and achieved good accuracy. Though a number of tools are available for the prediction of flexible length linear B-cell epitopes with reasonable accuracy, further improvement in the prediction performance is still expected. Thus, here we made an attempt to analyze the performance of machine learning approaches (MLA) with 18 different amino acid encoding schemes in the prediction of flexible length linear B-cell epitopes. We considered B-cell epitope sequences of variable lengths (11-56 amino acids) from well-established public resources. The performances of machine learning algorithms with the encoded epitope sequence datasets were evaluated. Besides, the feasible combinations of encoding schemes were also explored and analyzed. The results revealed that amino-acid composition (AC) and distribution component of composition-transition-distribution encoding schemes are suitable for heterogeneous epitope data, whereas amino-acid-anchoring-pair-composition (APC), dipeptide-composition and amino-acids-pair-propensity-scale (APP) are more appropriate for homogeneous data. Further, two combinations of peptide encoding schemes, i.e. APC + AC and APC + APP with random forest classifier were identified to have improved performance over the state-of-the-art tools for flexible length linear B-cell epitope prediction. The study also revealed better performance of random forest over other considered MLAs in the prediction of flexible length linear B-cell epitopes.

A critical challenge in genetic diagnostics is the assessment of genetic variants associated with diseases, specifically variants that fall out with canonical splice sites, by altering alternative splicing. Several computational methods have been developed to prioritize variants effect on splicing; however, performance evaluation of these methods is hampered by the lack of large-scale benchmark datasets. In this study, we employed a splicing-region-specific strategy to evaluate the performance of prediction methods based on eight independent datasets. Under most conditions, we found that dbscSNV-ADA performed better in the exonic region, S-CAP performed better in the core donor and acceptor regions, S-CAP and SpliceAI performed better in the extended acceptor region and MMSplice performed better in identifying variants that caused exon skipping. However, it should be noted that the performances of prediction methods varied widely under different datasets and splicing regions, and none of these methods showed the best overall performance with all datasets. To address this, we developed a new method, machine learning-based classification of splice sites variants (MLCsplice), to predict variants effect on splicing based on individual methods. We demonstrated that MLCsplice achieved stable and superior prediction performance compared with any individual method. To facilitate the identification of the splicing effect of variants, we provided precomputed MLCsplice scores for all possible splice sites variants across human protein-coding genes (http://39.105.51.3:8090/MLCsplice/). We believe that the performance of different individual methods under eight benchmark datasets will provide tentative guidance for appropriate method selection to prioritize candidate splice-disrupting variants, thereby increasing the genetic diagnostic yield.

Ensemble learning is a kind of machine learning method which can integrate multiple basic learners together and achieve higher accuracy. Recently, single machine learning methods have been established to predict survival for patients with cancer. However, it still lacked a robust ensemble learning model with high accuracy to pick out patients with high risks. To achieve this, we proposed a novel genetic algorithm-aided three-stage ensemble learning method (3S score) for survival prediction. During the process of constructing the 3S score, double training sets were used to avoid over-fitting; the gene-pairing method was applied to reduce batch effect; a genetic algorithm was employed to select the best basic learner combination. When used to predict the survival state of glioma patients, this model achieved the highest C-index (0.697) as well as area under the receiver operating characteristic curve (ROC-AUCs) (first year = 0.705, third year = 0.825 and fifth year = 0.839) in the combined test set (n = 1191), compared with 12 other baseline models. Furthermore, the 3S score can distinguish survival significantly in eight cohorts among the total of nine independent test cohorts (P < 0.05), achieving significant improvement of ROC-AUCs. Notably, ablation experiments demonstrated that the gene-pairing method, double training sets and genetic algorithm make sure the robustness and effectiveness of the 3S score. The performance exploration on pan-cancer showed that the 3S score has excellent ability on survival prediction in five kinds of cancers, which was verified by Cox regression, survival curves and ROC curves together. To enable its clinical adoption, we implemented the 3S score and other two clinical factors as an easy-to-use web tool for risk scoring and therapy stratification in glioma patients.

OBJECTIVES: Systemic Lupus Erythematosus is a complex autoimmune disease that leads to significant worsening of quality of life and mortality. Flares appear unpredictably during the disease course and therapies used are often only partially effective. These challenges are mainly due to the molecular heterogeneity of the disease, and in this context, personalized medicine-based approaches offer major promise. With this work we intended to advance in that direction by developing MyPROSLE, an omic-based analytical workflow for measuring the molecular portrait of individual patients to support clinicians in their therapeutic decisions. METHODS: Immunological gene-modules were used to represent the transcriptome of the patients. A dysregulation score for each gene-module was calculated at the patient level based on averaged z-scores. Almost 6100 Lupus and 750 healthy samples were used to analyze the association among dysregulation scores, clinical manifestations, prognosis, flare and remission events and response to Tabalumab. Machine learning-based classification models were built to predict around 100 different clinical parameters based on personalized dysregulation scores. RESULTS: MyPROSLE allows to molecularly summarize patients in 206 gene-modules, clustered into nine main lupus signatures. The combination of these modules revealed highly differentiated pathological mechanisms. We found that the dysregulation of certain gene-modules is strongly associated with specific clinical manifestations, the occurrence of relapses or the presence of long-term remission and drug response. Therefore, MyPROSLE may be used to accurately predict these clinical outcomes. CONCLUSIONS: MyPROSLE (https://myprosle.genyo.es) allows molecular characterization of individual Lupus patients and it extracts key molecular information to support more precise therapeutic decisions.

Drug-drug interactions (DDIs) prediction is a challenging task in drug development and clinical application. Due to the extremely large complete set of all possible DDIs, computer-aided DDIs prediction methods are getting lots of attention in the pharmaceutical industry and academia. However, most existing computational methods only use single perspective information and few of them conduct the task based on the biomedical knowledge graph (BKG), which can provide more detailed and comprehensive drug lateral side information flow. To this end, a deep learning framework, namely DeepLGF, is proposed to fully exploit BKG fusing local-global information to improve the performance of DDIs prediction. More specifically, DeepLGF first obtains chemical local information on drug sequence semantics through a natural language processing algorithm. Then a model of BFGNN based on graph neural network is proposed to extract biological local information on drug through learning embedding vector from different biological functional spaces. The global feature information is extracted from the BKG by our knowledge graph embedding method. In DeepLGF, for fusing local-global features well, we designed four aggregating methods to explore the most suitable ones. Finally, the advanced fusing feature vectors are fed into deep neural network to train and predict. To evaluate the prediction performance of DeepLGF, we tested our method in three prediction tasks and compared it with state-of-the-art models. In addition, case studies of three cancer-related and COVID-19-related drugs further demonstrated DeepLGF's superior ability for potential DDIs prediction. The webserver of the DeepLGF predictor is freely available at http://120.77.11.78/DeepLGF/.

Maintaining duplicate germplasms in genebanks hampers effective conservation and utilization of genebank resources. The redundant germplasm adds to the cost of germplasm conservation by requiring a large proportion of the genebank financial resources towards conservation rather than enriching the diversity. Besides, genome-wide-association analysis using an association panel with over-represented germplasms can be biased resulting in spurious marker-trait associations. The conventional methods of germplasm duplicate removal using passport information suffer from incomplete or missing passport information and data handling errors at various stages of germplasm enrichment. This limitation is less likely in the case of genotypic data. Therefore, we developed a web-based tool, Germplasm Duplicate Identification and Removal Tool (G-DIRT), which allows germplasm duplicate identification based on identity-by-state analysis using single-nucleotide polymorphism genotyping information along with pre-processing of genotypic data. A homozygous genotypic difference threshold of 0.1% for germplasm duplicates has been determined using tetraploid wheat genotypic data with 94.97% of accuracy. Based on the genotypic difference, the tool also builds a dendrogram that can visually depict the relationship between genotypes. To overcome the constraint of high-dimensional genotypic data, an offline version of G-DIRT in the interface of R has also been developed. The G-DIRT is expected to help genebank curators, breeders and other researchers across the world in identifying germplasm duplicates from the global genebank collections by only using the easily sharable genotypic data instead of physically exchanging the seeds or propagating materials. The web server will complement the existing methods of germplasm duplicate identification based on passport or phenotypic information being freely accessible at http://webtools.nbpgr.ernet.in/gdirt/.

A newly invented post-translational modification (PTM), phosphoglycerylation, has shown its essential role in the construction and functional properties of proteins and dangerous human diseases. Hence, it is very urgent to know about the molecular mechanism behind the phosphoglycerylation process to develop the drugs for related diseases. But accurately identifying of phosphoglycerylation site from a protein sequence in a laboratory is a very difficult and challenging task. Hence, the construction of an efficient computation model is greatly sought for this purpose. A little number of computational models are currently available for identifying the phosphoglycerylation sites, which are not able to reach their prediction capability at a satisfactory level. Therefore, an effective predictor named PLP_FS has been designed and constructed to identify phosphoglycerylation sites in this study. For the training purpose, an optimal number of feature sets was obtained by fusion of multiple F_Score feature selection techniques from the features generated by three types of sequence-based feature extraction methods and fitted with the support vector machine classification technique to the prediction model. On the other hand, the k-neighbor near cleaning and SMOTE methods were also implemented to balance the benchmark dataset. The suggested model in 10-fold cross-validation obtained an accuracy of 99.22%, a sensitivity of 98.17% and a specificity of 99.75% according to the experimental findings, which are better than other currently available predictors for accurately identifying the phosphoglycerylation sites.

A knowledge-based grouping of genes into pathways or functional units is essential for describing and understanding cellular complexity. However, it is not always clear a priori how and at what level of specificity functionally interconnected genes should be partitioned into pathways, for a given application. Here, we assess and compare nine existing and two conceptually novel functional classification systems, with respect to their discovery power and generality in gene set enrichment testing. We base our assessment on a collection of nearly 2000 functional genomics datasets provided by users of the STRING database. With these real-life and diverse queries, we assess which systems typically provide the most specific and complete enrichment results. We find many structural and performance differences between classification systems. Overall, the well-established, hierarchically organized pathway annotation systems yield the best enrichment performance, despite covering substantial parts of the human genome in general terms only. On the other hand, the more recent unsupervised annotation systems perform strongest in understudied areas and organisms, and in detecting more specific pathways, albeit with less informative labels.

The rapid development of genomic high-throughput sequencing has identified a large number of DNA regulatory elements with abundant epigenetics markers, which promotes the rapid accumulation of functional genomic region data. The comprehensively understanding and research of human functional genomic regions is still a relatively urgent work at present. However, the existing analysis tools lack extensive annotation and enrichment analytical abilities for these regions. Here, we designed a novel software, Genomic Region sets Enrichment Analysis Platform (GREAP), which provides comprehensive region annotation and enrichment analysis capabilities. Currently, GREAP supports 85 370 genomic region reference sets, which cover 634 681 107 regions across 11 different data types, including super enhancers, transcription factors, accessible chromatins, etc. GREAP provides widespread annotation and enrichment analysis of genomic regions. To reflect the significance of enrichment analysis, we used the hypergeometric test and also provided a Locus Overlap Analysis. In summary, GREAP is a powerful platform that provides many types of genomic region sets for users and supports genomic region annotations and enrichment analyses. In addition, we developed a customizable genome browser containing >400 000 000 customizable tracks for visualization. The platform is freely available at http://www.liclab.net/Greap/view/index.

Metals are present in >30% of proteins found in nature and assist them to perform important biological functions, including storage, transport, signal transduction and enzymatic activity. Traditional and experimental techniques for metal-binding site prediction are usually costly and time-consuming, making computational tools that can assist in these predictions of significant importance. Here we present Genetic Active Site Search (GASS)-Metal, a new method for protein metal-binding site prediction. The method relies on a parallel genetic algorithm to find candidate metal-binding sites that are structurally similar to curated templates from M-CSA and MetalPDB. GASS-Metal was thoroughly validated using homologous proteins and conservative mutations of residues, showing a robust performance. The ability of GASS-Metal to identify metal-binding sites was also compared with state-of-the-art methods, outperforming similar methods and achieving an MCC of up to 0.57 and detecting up to 96.1% of the sites correctly. GASS-Metal is freely available at https://gassmetal.unifei.edu.br. The GASS-Metal source code is available at https://github.com/sandroizidoro/gassmetal-local.

Multiple primary tumor (MPT) is a special and rare cancer type, defined as more than two primary tumors presenting at the diagnosis in a single patient. The molecular characteristics and tumorigenesis of MPT remain unclear due to insufficient approaches. Here, we present MPTevol, a practical computational framework for comprehensively exploring the MPT from multiregion sequencing (MRS) experiments. To verify the utility of MPTevol, we performed whole-exome MRS for 33 samples of a rare patient with triple-primary tumors and three metastatic sites and systematically investigated clonal dynamics and metastatic routines. MPTevol assists in comparing genomic profiles across samples, detecting clonal evolutionary history and metastatic routines and quantifying the metastatic history. All triple-primary tumors were independent origins and their genomic characteristics were consistent with corresponding sporadic tumors, strongly supporting their independent tumorigenesis. We further showed two independent early monoclonal seeding events for the metastases in the ovary and uterus. We revealed that two ovarian metastases were disseminated from the same subclone of the primary tumor through undergoing whole-genome doubling processes, suggesting metastases-to-metastases seeding occurred when tumors had similar microenvironments. Surprisingly, according to the metastasis timing model of MPTevol, we found that primary tumors of about 0.058-0.124 cm diameter have been disseminating to distant organs, which is much earlier than conventional clinical views. We developed MPT-specialized analysis framework MPTevol and demonstrated its utility in explicitly resolving clonal evolutionary history and metastatic seeding routines with a rare MPT case. MPTevol is implemented in R and is available at https://github.com/qingjian1991/MPTevol under the GPL v3 license.

Antibodies are essential to life, and knowing their structures can facilitate the understanding of antibody-antigen recognition mechanisms. Precise antibody structure prediction has been a core challenge for a prolonged period, especially the accuracy of H3 loop prediction. Despite recent progress, existing methods cannot achieve atomic accuracy, especially when the homologous structures required for these methods are not available. Recently, RoseTTAFold, a deep learning-based algorithm, has shown remarkable breakthroughs in predicting the 3D structures of proteins. To assess the antibody modeling ability of RoseTTAFold, we first retrieved the sequences of 30 antibodies as the test set and used RoseTTAFold to model their 3D structures. We then compared the models constructed by RoseTTAFold with those of SWISS-MODEL in a different way, in which we stratified Global Model Quality Estimate (GMQE) into three different ranges. The results indicated that RoseTTAFold could achieve results similar to SWISS-MODEL in modeling most CDR loops, especially the templates with a GMQE score under 0.8. In addition, we also compared the structures modeled by RoseTTAFold, SWISS-MODEL and ABodyBuilder. In brief, RoseTTAFold could accurately predict 3D structures of antibodies, but its accuracy was not as good as the other two methods. However, RoseTTAFold exhibited better accuracy for modeling H3 loop than ABodyBuilder and was comparable to SWISS-MODEL. Finally, we discussed the limitations and potential improvements of the current RoseTTAFold, which may help to further the accuracy of RoseTTAFold's antibody modeling.

Pharmacological chaperones are chemical compounds able to bind proteins and stabilize them against denaturation and following degradation. Some pharmacological chaperones have been approved, or are under investigation, for the treatment of rare inborn errors of metabolism, caused by genetic mutations that often can destabilize the structure of the wild-type proteins expressed by that gene. Given that, for rare diseases, there is a general lack of pharmacological treatments, many expectations are poured out on this type of compounds. However, their discovery is not straightforward. In this review, we would like to focus on the computational methods that can assist and accelerate the search for these compounds, showing also examples in which these methods were successfully applied for the discovery of promising molecules belonging to this new category of pharmacologically active compounds.

Yuan et al. recently described an independent evaluation of several phenotype-driven gene prioritization methods for Mendelian disease on two separate, clinical datasets. Although they attempted to use default settings for each tool, we describe three key differences from those we currently recommend for our Exomiser and PhenIX tools. These influence how variant frequency, quality and predicted pathogenicity are used for filtering and prioritization. We propose that these differences account for much of the discrepancy in performance between that reported by them (15-26% diagnoses ranked top by Exomiser) and previously published reports by us and others (72-77%). On a set of 161 singleton samples, we show using these settings increases performance from 34% to 72% and suggest a reassessment of Exomiser and PhenIX on their datasets using these would show a similar uplift.

Predicting the native or near-native binding pose of a small molecule within a protein binding pocket is an extremely important task in structure-based drug design, especially in the hit-to-lead and lead optimization phases. In this study, fastDRH, a free and open accessed web server, was developed to predict and analyze protein-ligand complex structures. In fastDRH server, AutoDock Vina and AutoDock-GPU docking engines, structure-truncated MM/PB(GB)SA free energy calculation procedures and multiple poses based per-residue energy decomposition analysis were well integrated into a user-friendly and multifunctional online platform. Benefit from the modular architecture, users can flexibly use one or more of three features, including molecular docking, docking pose rescoring and hotspot residue prediction, to obtain the key information clearly based on a result analysis panel supported by 3Dmol.js and Apache ECharts. In terms of protein-ligand binding mode prediction, the integrated structure-truncated MM/PB(GB)SA rescoring procedures exhibit a success rate of >80% in benchmark, which is much better than the AutoDock Vina (~70%). For hotspot residue identification, our multiple poses based per-residue energy decomposition analysis strategy is a more reliable solution than the one using only a single pose, and the performance of our solution has been experimentally validated in several drug discovery projects. To summarize, the fastDRH server is a useful tool for predicting the ligand binding mode and the hotspot residue of protein for ligand binding. The fastDRH server is accessible free of charge at http://cadd.zju.edu.cn/fastdrh/.

Complex diseases progression can be generally divided into three states, which are normal state, predisease/critical state and disease state. The sudden deterioration of diseases can be viewed as a bifurcation or a critical transition. Therefore, hunting for the tipping point or critical state is of great importance to prevent the disease deterioration. However, it is still a challenging task to detect the critical states of complex diseases with high-dimensional data, especially based on an individual. In this study, we develop a new method based on network fluctuation of molecules, temporal network flow entropy (TNFE) or temporal differential network flow entropy, to detect the critical states of complex diseases on the basis of each individual. By applying this method to a simulated dataset and six real diseases, including respiratory viral infections and tumors with four time-course and two stage-course high-dimensional omics datasets, the critical states before deterioration were detected and their dynamic network biomarkers were identified successfully. The results on the simulated dataset indicate that the TNFE method is robust under different noise strengths, and is also superior to the existing methods on detecting the critical states. Moreover, the analysis on the real datasets demonstrated the effectiveness of TNFE for providing early-warning signals on various diseases. In addition, we also predicted disease deterioration risk and identified drug targets for cancers based on stage-wise data.

The association between the compositions of microbial communities and various host phenotypes is an important research topic. Microbiome association research addresses multiple domains, such as human disease and diet. Statistical methods for testing microbiome-phenotype associations have been studied recently to determine their ability to assess longitudinal microbiome data. However, existing methods fail to detect sparse association signals in longitudinal microbiome data. In this paper, we developed a novel method, namely aGEEMIHC, which is a data-driven adaptive microbiome higher criticism analysis based on generalized estimating equations to detect sparse microbial association signals from longitudinal microbiome data. aGEEMiHC adopts generalized estimating equations framework that fully considers the correlation among different observations from the same subject in longitudinal data. To be robust to diverse correlation structures for longitudinal data, aGEEMiHC integrates multiple microbiome higher criticism analyses based on generalized estimating equations with different working correlation structures. Extensive simulation experiments demonstrate that aGEEMiHC can control the type I error correctly and achieve superior performance according to a statistical power comparison. We also applied it to longitudinal microbiome data with various types of host phenotypes to demonstrate the stability of our method. aGEEMiHC is also utilized for real longitudinal microbiome data, and we found a significant association between the gut microbiome and Crohn's disease. In addition, our method ranks the significant factors associated with the host phenotype to provide potential biomarkers.

Author information: (1)Academy of Statistics and Interdisciplinary Sciences, KLATASDS-MOE, East China Normal University, Shanghai, China. (2)Department of Mathematics, The Hong Kong University of Science and Technology, Hong Kong SAR, China. (3)Department of Statistics, The Chinese University of Hong Kong, Hong Kong SAR, China. (4)Shenzhen Research Institute of Big Data, Shenzhen, China. (5)Pazhou Lab, Guangzhou, China. (6)Division of Life Science, The Hong Kong University of Science and Technology, Hong Kong SAR, China. (7)Department of Chemical and Biological Engineering, The Hong Kong University of Science and Technology, Hong Kong SAR, China.

Understanding the biological functions of molecules in specific human tissues or cell types is crucial for gaining insights into human physiology and disease. To address this issue, it is essential to systematically uncover associations among multilevel elements consisting of disease phenotypes, tissues, cell types and molecules, which could pose a challenge because of their heterogeneity and incompleteness. To address this challenge, we describe a new methodological framework, called Graph Local InfoMax (GLIM), based on a human multilevel network (HMLN) that we established by introducing multiple tissues and cell types on top of molecular networks. GLIM can systematically mine the potential relationships between multilevel elements by embedding the features of the HMLN through contrastive learning. Our simulation results demonstrated that GLIM consistently outperforms other state-of-the-art algorithms in disease gene prediction. Moreover, GLIM was also successfully used to infer cell markers and rewire intercellular and molecular interactions in the context of specific tissues or diseases. As a typical case, the tissue-cell-molecule network underlying gastritis and gastric cancer was first uncovered by GLIM, providing systematic insights into the mechanism underlying the occurrence and development of gastric cancer. Overall, our constructed methodological framework has the potential to systematically uncover complex disease mechanisms and mine high-quality relationships among phenotypical, tissue, cellular and molecular elements.

MOTIVATION: In recent years, a large number of biological experiments have strongly shown that miRNAs play an important role in understanding disease pathogenesis. The discovery of miRNA-disease associations is beneficial for disease diagnosis and treatment. Since inferring these associations through biological experiments is time-consuming and expensive, researchers have sought to identify the associations utilizing computational approaches. Graph Convolutional Networks (GCNs), which exhibit excellent performance in link prediction problems, have been successfully used in miRNA-disease association prediction. However, GCNs only consider 1st-order neighborhood information at one layer but fail to capture information from high-order neighbors to learn miRNA and disease representations through information propagation. Therefore, how to aggregate information from high-order neighborhood effectively in an explicit way is still challenging. RESULTS: To address such a challenge, we propose a novel method called mixed neighborhood information for miRNA-disease association (MINIMDA), which could fuse mixed high-order neighborhood information of miRNAs and diseases in multimodal networks. First, MINIMDA constructs the integrated miRNA similarity network and integrated disease similarity network respectively with their multisource information. Then, the embedding representations of miRNAs and diseases are obtained by fusing mixed high-order neighborhood information from multimodal network which are the integrated miRNA similarity network, integrated disease similarity network and the miRNA-disease association networks. Finally, we concentrate the multimodal embedding representations of miRNAs and diseases and feed them into the multilayer perceptron (MLP) to predict their underlying associations. Extensive experimental results show that MINIMDA is superior to other state-of-the-art methods overall. Moreover, the outstanding performance on case studies for esophageal cancer, colon tumor and lung cancer further demonstrates the effectiveness of MINIMDA. AVAILABILITY AND IMPLEMENTATION: https://github.com/chengxu123/MINIMDA and http://120.79.173.96/.

High-throughput genomic technologies are increasingly used in personalized cancer medicine. However, computational tools to maximize the use of scarce tissues combining distinct molecular layers are needed. Here we present a refined strategy, based on the R-package 'conumee', to better predict somatic copy number alterations (SCNA) from deoxyribonucleic acid (DNA) methylation arrays. Our approach, termed hereafter as 'conumee-KCN', improves SCNA prediction by incorporating tumor purity and dynamic thresholding. We trained our algorithm using paired DNA methylation and SNP Array 6.0 data from The Cancer Genome Atlas samples and confirmed its performance in cancer cell lines. Most importantly, the application of our approach in cancers of unknown primary identified amplified potentially actionable targets that were experimentally validated by Fluorescence in situ hybridization and immunostaining, reaching 100% specificity and 93.3% sensitivity.

Molecular signatures have been excessively reported for diagnosis of many cancers during the last 20 years. However, false-positive signatures are always found using statistical methods or machine learning approaches, and that makes subsequent biological experiments fail. Therefore, signature discovery has gradually become a non-mainstream work in bioinformatics. Actually, there are three critical weaknesses that make the identified signature unreliable. First of all, a signature is wrongly thought to be a gene set, each component of which keeps differential expressions between or among sample groups. Second, there may be many false-positive genes expressed differentially found, even if samples derived from cancer or normal group can be separated in one-dimensional space. Third, cross-platform validation results of a discovered signature are always poor. In order to solve these problems, we propose a new feature selection framework based on ensemble classification to discover signatures for cancer diagnosis. Meanwhile, a procedure for data transform among different expression profiles across different platforms is also designed. Signatures are found on simulation and real data representing different carcinomas across different platforms. Besides, false positives are suppressed. The experimental results demonstrate the effectiveness of our method.

The dynamics of complex diseases are not always smooth; they are occasionally abrupt, i.e. there is a critical state transition or tipping point at which the disease undergoes a sudden qualitative shift. There are generally a few significant differences in the critical state in terms of gene expressions or other static measurements, which may lead to the failure of traditional differential expression-based biomarkers to identify such a tipping point. In this study, we propose a computational method, the direct interaction network-based divergence, to detect the critical state of complex diseases by exploiting the dynamic changes in multivariable distributions inferred from observable samples and local biomolecular direct interaction networks. Such a method is model-free and applicable to both bulk and single-cell expression data. Our approach was validated by successfully identifying the tipping point just before the occurrence of a critical transition for both a simulated data set and seven real data sets, including those from The Cancer Genome Atlas and two single-cell RNA-sequencing data sets of cell differentiation. Functional and pathway enrichment analyses also validated the computational results from the perspectives of both molecules and networks.

Sex-biased gene flow has been common in the demographic history of modern humans. However, the lack of sophisticated methods for delineating the detailed sex-biased admixture process prevents insights into complex admixture history and thus our understanding of the evolutionary mechanisms of genetic diversity. Here, we present a novel algorithm, MultiWaverX, for modeling complex admixture history with sex-biased gene flow. Systematic simulations showed that MultiWaverX is a powerful tool for modeling complex admixture history and inferring sex-biased gene flow. Application of MultiWaverX to empirical data of 17 typical admixed populations in America, Central Asia, and the Middle East revealed sex-biased admixture histories that were largely consistent with the historical records. Notably, fine-scale admixture process reconstruction enabled us to recognize latent sex-biased gene flow in certain populations that would likely be overlooked by much of the routine analysis with commonly used methods. An outstanding example in the real world is the Kazakh population that experienced complex admixture with sex-biased gene flow but in which the overall signature has been canceled due to biased gene flow from an opposite direction.

Differential expression (DE) gene detection in single-cell ribonucleic acid (RNA)-sequencing (scRNA-seq) data is a key step to understand the biological question investigated. Filtering genes is suggested to improve the performance of DE methods, but the influence of filtering genes has not been demonstrated. Furthermore, the optimal methods for different scRNA-seq datasets are divergent, and different datasets should benefit from data-specific DE gene detection strategies. However, existing tools did not take gene filtering into consideration. There is a lack of metrics for evaluating the optimal method on experimental datasets. Based on two new metrics, we propose single-cell Consensus Optimization of Differentially Expressed gene detection, an R package to automatically optimize DE gene detection for each experimental scRNA-seq dataset.

Many statistical methods for pathway analysis have been used to identify pathways associated with the disease along with biological factors such as genes and proteins. However, most pathway analysis methods neglect the complex nonlinear relationship between biological factors and pathways. In this study, we propose a Deep-learning pathway analysis using Hierarchical structured CoMponent models (DeepHisCoM) that utilize deep learning to consider a nonlinear complex contribution of biological factors to pathways by constructing a multilayered model which accounts for hierarchical biological structure. Through simulation studies, DeepHisCoM was shown to have a higher power in the nonlinear pathway effect and comparable power for the linear pathway effect when compared to the conventional pathway methods. Application to hepatocellular carcinoma (HCC) omics datasets, including metabolomic, transcriptomic and metagenomic datasets, demonstrated that DeepHisCoM successfully identified three well-known pathways that are highly associated with HCC, such as lysine degradation, valine, leucine and isoleucine biosynthesis and phenylalanine, tyrosine and tryptophan. Application to the coronavirus disease-2019 (COVID-19) single-nucleotide polymorphism (SNP) dataset also showed that DeepHisCoM identified four pathways that are highly associated with the severity of COVID-19, such as mitogen-activated protein kinase (MAPK) signaling pathway, gonadotropin-releasing hormone (GnRH) signaling pathway, hypertrophic cardiomyopathy and dilated cardiomyopathy. Codes are available at https://github.com/chanwoo-park-official/DeepHisCoM.

Distinguishing pathogenic variants from non-pathogenic ones remains a major challenge in clinical genetic testing of primary immunodeficiency (PID) patients. Most of the existing mutation pathogenicity prediction tools treat all mutations as homogeneous entities, ignoring the differences in characteristics of different genes, and use the same model for genes in different diseases. In this study, we developed a single nucleotide variant (SNV) pathogenicity prediction tool, Variant Impact Predictor for PIDs (VIPPID; https://mylab.shinyapps.io/VIPPID/), which was tailored for PIDs genes and used a specific model for each of the most prevalent PID known genes. It employed a Conditional Inference Forest model and utilized information of 85 features of SNVs and scores from 20 existing prediction tools. Evaluation of VIPPID showed that it had superior performance (area under the curve = 0.91) over non-specific conventional tools. In addition, we also showed that the gene-specific model outperformed the non-gene-specific models. Our study demonstrated that disease-specific and gene-specific models can improve SNV pathogenicity prediction performance. This observation supports the notion that each feature of mutations in the model can be potentially used, in a new algorithm, to investigate the characteristics and function of the encoded proteins.

The spatial and temporal domain of a gene's expression can range from ubiquitous to highly specific. Quantifying the degree to which this expression is unique to a specific tissue or developmental timepoint can provide insight into the etiology of genetic diseases. However, quantifying specificity remains challenging as measures of specificity are sensitive to similarity between samples in the sample set. For example, in the Gene-Tissue Expression project (GTEx), brain subregions are overrepresented at 13 of 54 (24%) unique tissues sampled. In this dataset, existing specificity measures have a decreased ability to identify genes specific to the brain relative to other organs. To solve this problem, we leverage sample similarity information to weight samples such that overrepresented tissues do not have an outsized effect on specificity estimates. We test this reweighting procedure on 4 measures of specificity, Z-score, Tau, Tsi and Gini, in the GTEx data and in single cell datasets for zebrafish and mouse. For all of these measures, incorporating sample similarity information to weight samples results in greater stability of sets of genes called as specific and decreases the overall variance in the change of specificity estimates as sample sets become more unbalanced. Furthermore, the genes with the largest improvement in their specificity estimate's stability are those with functions related to the overrepresented sample types. Our results demonstrate that incorporating similarity information improves specificity estimates' stability to the choice of the sample set used to define the transcriptome, providing more robust and reproducible measures of specificity for downstream analyses.

BACKGROUND: Coronavirus disease 2019 (COVID-19) has spurred a boom in uncovering repurposable existing drugs. Drug repurposing is a strategy for identifying new uses for approved or investigational drugs that are outside the scope of the original medical indication. MOTIVATION: Current works of drug repurposing for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) are mostly limited to only focusing on chemical medicines, analysis of single drug targeting single SARS-CoV-2 protein, one-size-fits-all strategy using the same treatment (same drug) for different infected stages of SARS-CoV-2. To dilute these issues, we initially set the research focusing on herbal medicines. We then proposed a heterogeneous graph embedding method to signaled candidate repurposing herbs for each SARS-CoV-2 protein, and employed the variational graph convolutional network approach to recommend the precision herb combinations as the potential candidate treatments against the specific infected stage. METHOD: We initially employed the virtual screening method to construct the 'Herb-Compound' and 'Compound-Protein' docking graph based on 480 herbal medicines, 12,735 associated chemical compounds and 24 SARS-CoV-2 proteins. Sequentially, the 'Herb-Compound-Protein' heterogeneous network was constructed by means of the metapath-based embedding approach. We then proposed the heterogeneous-information-network-based graph embedding method to generate the candidate ranking lists of herbs that target structural, nonstructural and accessory SARS-CoV-2 proteins, individually. To obtain precision synthetic effective treatments forvarious COVID-19 infected stages, we employed the variational graph convolutional network method to generate candidate herb combinations as the recommended therapeutic therapies. RESULTS: There were 24 ranking lists, each containing top-10 herbs, targeting 24 SARS-CoV-2 proteins correspondingly, and 20 herb combinations were generated as the candidate-specific treatment to target the four infected stages. The code and supplementary materials are freely available at https://github.com/fanyang-AI/TCM-COVID19.

Human Leukocyte Antigen (HLA) is a type of molecule residing on the surfaces of most human cells and exerts an essential role in the immune system responding to the invasive items. The T cell antigen receptors may recognize the HLA-peptide complexes on the surfaces of cancer cells and destroy these cancer cells through toxic T lymphocytes. The computational determination of HLA-binding peptides will facilitate the rapid development of cancer immunotherapies. This study hypothesized that the natural language processing-encoded peptide features may be further enriched by another deep neural network. The hypothesis was tested with the Bi-directional Long Short-Term Memory-extracted features from the pretrained Protein Bidirectional Encoder Representations from Transformers-encoded features of the class I HLA (HLA-I)-binding peptides. The experimental data showed that our proposed HLAB feature engineering algorithm outperformed the existing ones in detecting the HLA-I-binding peptides. The extensive evaluation data show that the proposed HLAB algorithm outperforms all the seven existing studies on predicting the peptides binding to the HLA-A*01:01 allele in AUC and achieves the best average AUC values on the six out of the seven k-mers (k=8,9,...,14, respectively represent the prediction task of a polypeptide consisting of k amino acids) except for the 9-mer prediction tasks. The source code and the fine-tuned feature extraction models are available at http://www.healthinformaticslab.org/supp/resources.php.

The development of single-cell RNA-sequencing (scRNA-seq) technologies has offered insights into complex biological systems at the single-cell resolution. In particular, these techniques facilitate the identifications of genes showing cell-type-specific differential expressions (DE). In this paper, we introduce MARBLES, a novel statistical model for cross-condition DE gene detection from scRNA-seq data. MARBLES employs a Markov Random Field model to borrow information across similar cell types and utilizes cell-type-specific pseudobulk count to account for sample-level variability. Our simulation results showed that MARBLES is more powerful than existing methods to detect DE genes with an appropriate control of false positive rate. Applications of MARBLES to real data identified novel disease-related DE genes and biological pathways from both a single-cell lipopolysaccharide mouse dataset with 24 381 cells and 11 076 genes and a Parkinson's disease human data set with 76 212 cells and 15 891 genes. Overall, MARBLES is a powerful tool to identify cell-type-specific DE genes across conditions from scRNA-seq data.

The ubiquitous dropout problem in single-cell RNA sequencing technology causes a large amount of data noise in the gene expression profile. For this reason, we propose an evolutionary sparse imputation (ESI) algorithm for single-cell transcriptomes, which constructs a sparse representation model based on gene regulation relationships between cells. To solve this model, we design an optimization framework based on nondominated sorting genetics. This framework takes into account the topological relationship between cells and the variety of gene expression to iteratively search the global optimal solution, thereby learning the Pareto optimal cell-cell affinity matrix. Finally, we use the learned sparse relationship model between cells to improve data quality and reduce data noise. In simulated datasets, scESI performed significantly better than benchmark methods with various metrics. By applying scESI to real scRNA-seq datasets, we discovered scESI can not only further classify the cell types and separate cells in visualization successfully but also improve the performance in reconstructing trajectories differentiation and identifying differentially expressed genes. In addition, scESI successfully recovered the expression trends of marker genes in stem cell differentiation and can discover new cell types and putative pathways regulating biological processes.