Long noncoding ribonucleic acids (RNAs; lncRNAs) have been associated with cancer immunity regulation. However, the roles of immune cell-specific lncRNAs in glioblastoma (GBM) remain largely unknown. In this study, a novel computational framework was constructed to screen the tumor-infiltrating immune cell-associated lncRNAs (TIIClnc) for developing TIIClnc signature by integratively analyzing the transcriptome data of purified immune cells, GBM cell lines and bulk GBM tissues using six machine learning algorithms. As a result, TIIClnc signature could distinguish survival outcomes of GBM patients across four independent datasets, including the Xiangya in-house dataset, and more importantly, showed superior performance than 95 previously established signatures in gliomas. TIIClnc signature was revealed to be an indicator of the infiltration level of immune cells and predicted the response outcomes of immunotherapy. The positive correlation between TIIClnc signature and CD8, PD-1 and PD-L1 was verified in the Xiangya in-house dataset. As a newly demonstrated predictive biomarker, the TIIClnc signature enabled a more precise selection of the GBM population who would benefit from immunotherapy and should be validated and applied in the near future.

Biomedical data preprocessing and efficient computing can be as important as the statistical methods used to fit the data; data processing needs to consider application scenarios, data acquisition and individual rights and interests. We review common principles, knowledge and methods of integrated research according to the whole-pipeline processing mechanism diverse, coherent, sharing, auditable and ecological. First, neuromorphic and native algorithms integrate diverse datasets, providing linear scalability and high visualization. Second, the choice mechanism of different preprocessing, analysis and transaction methods from raw to neuromorphic was summarized on the node and coordinator platforms. Third, combination of node, network, cloud, edge, swarm and graph builds an ecosystem of cohort integrated research and clinical diagnosis and treatment. Looking forward, it is vital to simultaneously combine deep computing, mass data storage and massively parallel communication.

In common medical procedures, the time-consuming and expensive nature of obtaining test results plagues doctors and patients. Digital pathology research allows using computational technologies to manage data, presenting an opportunity to improve the efficiency of diagnosis and treatment. Artificial intelligence (AI) has a great advantage in the data analytics phase. Extensive research has shown that AI algorithms can produce more up-to-date and standardized conclusions for whole slide images. In conjunction with the development of high-throughput sequencing technologies, algorithms can integrate and analyze data from multiple modalities to explore the correspondence between morphological features and gene expression. This review investigates using the most popular image data, hematoxylin-eosin stained tissue slide images, to find a strategic solution for the imbalance of healthcare resources. The article focuses on the role that the development of deep learning technology has in assisting doctors' work and discusses the opportunities and challenges of AI.

The human genome is marked by several singular and combinatorial histone modifications that shape the different states of chromatin and its three-dimensional organization. Genome-wide mapping of these marks as well as histone variants and open chromatin regions is commonly carried out via profiling DNA-protein binding or via chromatin accessibility methods. After the generation of epigenomic datasets in a cell type, statistical models can be used to annotate the noncoding regions of DNA and infer the combinatorial histone marks or chromatin states (CS). These methods involve partitioning the genome and labeling individual segments based on their CS patterns. Chromatin labels enable the systematic discovery of genomic function and activity and can label the gene body, promoters or enhancers without using other genomic maps. CSs are dynamic and change under different cell conditions, such as in normal, preneoplastic or tumor cells. This review aims to explore the available computational tools that have been developed to capture CS alterations under two or more cellular conditions.

Emerging evidence indicates that circular RNAs (circRNAs) can provide new insights and potential therapeutic targets for disease diagnosis and treatment. However, traditional biological experiments are expensive and time-consuming. Recently, deep learning with a more powerful ability for representation learning enables it to be a promising technology for predicting disease-associated circRNAs. In this review, we mainly introduce the most popular databases related to circRNA, and summarize three types of deep learning-based circRNA-disease associations prediction methods: feature-generation-based, type-discrimination and hybrid-based methods. We further evaluate seven representative models on benchmark with ground truth for both balance and imbalance classification tasks. In addition, we discuss the advantages and limitations of each type of method and highlight suggested applications for future research.

As space exploration programs progress, manned space missions will become more frequent and farther away from Earth, putting a greater emphasis on astronaut health. Through the collaborative efforts of researchers from various countries, the effect of the space environment factors on living systems is gradually being uncovered. Although a large number of interconnected research findings have been produced, their connection seems to be confused, and many unknown effects are left to be discovered. Simultaneously, several valuable data resources have emerged, accumulating data measuring biological effects in space that can be used to further investigate the unknown biological adaptations. In this review, the previous findings and their correlations are sorted out to facilitate the understanding of biological adaptations to space and the design of countermeasures. The biological effect measurement methods/data types are also organized to provide references for experimental design and data analysis. To aid deeper exploration of the data resources, we summarized common characteristics of the data generated from longitudinal experiments, outlined challenges or caveats in data analysis and provided corresponding solutions by recommending bioinformatics strategies and available models/tools.

Microbial community classification enables identification of putative type and source of the microbial community, thus facilitating a better understanding of how the taxonomic and functional structure were developed and maintained. However, previous classification models required a trade-off between speed and accuracy, and faced difficulties to be customized for a variety of contexts, especially less studied contexts. Here, we introduced EXPERT based on transfer learning that enabled the classification model to be adaptable in multiple contexts, with both high efficiency and accuracy. More importantly, we demonstrated that transfer learning can facilitate microbial community classification in diverse contexts, such as classification of microbial communities for multiple diseases with limited number of samples, as well as prediction of the changes in gut microbiome across successive stages of colorectal cancer. Broadly, EXPERT enables accurate and context-aware customized microbial community classification, and potentiates novel microbial knowledge discovery.

Biological networks are important for the analysis of human diseases, which summarize the regulatory interactions and other relationships between different molecules. Understanding and constructing networks for molecules, such as DNA, RNA and proteins, can help elucidate the mechanisms of complex biological systems. The Gaussian Graphical Models (GGMs) are popular tools for the estimation of biological networks. Nonetheless, reconstructing GGMs from high-dimensional datasets is still challenging. The current methods cannot handle the sparsity and high-dimensionality issues arising from datasets very well. Here, we developed a new GGM, called the GR2D2 (Graphical $R^2$-induced Dirichlet Decomposition) model, based on the R2D2 priors for linear models. Besides, we provided a data-augmented block Gibbs sampler algorithm. The R code is available at https://github.com/RavenGan/GR2D2. The GR2D2 estimator shows superior performance in estimating the precision matrices compared with the existing techniques in various simulation settings. When the true precision matrix is sparse and of high dimension, the GR2D2 provides the estimates with smallest information divergence from the underlying truth. We also compare the GR2D2 estimator with the graphical horseshoe estimator in five cancer RNA-seq gene expression datasets grouped by three cancer types. Our results show that GR2D2 successfully identifies common cancer pathways and cancer-specific pathways for each dataset.

Gene-based transcriptome analysis, such as differential expression analysis, can identify the key factors causing disease production, cell differentiation and other biological processes. However, this is not enough because basic life activities are mainly driven by the interactions between genes. Although there have been already many differential network inference methods for identifying the differential gene interactions, currently, most studies still only use the information of nodes in the network for downstream analyses. To investigate the insight into differential gene interactions, we should perform interaction-based transcriptome analysis (IBTA) instead of gene-based analysis after obtaining the differential networks. In this paper, we illustrated a workflow of IBTA by developing a Co-hub Differential Network inference (CDN) algorithm, and a novel interaction-based metric, pivot APC2. We confirmed the superior performance of CDN through simulation experiments compared with other popular differential network inference algorithms. Furthermore, three case studies are given using colorectal cancer, COVID-19 and triple-negative breast cancer datasets to demonstrate the ability of our interaction-based analytical process to uncover causative mechanisms.

For many high-dimensional genomic and epigenomic datasets, the outcome of interest is ordinal. While these ordinal outcomes are often thought of as the observed cutpoints of some latent continuous variable, some ordinal outcomes are truly discrete and are comprised of the subjective combination of several factors. The nonlinear stereotype logistic model, which does not assume proportional odds, was developed for these 'assessed' ordinal variables. It has previously been extended to the frequentist high-dimensional feature selection setting, but the Bayesian framework provides some distinct advantages in terms of simultaneous uncertainty quantification and variable selection. Here, we review the stereotype model and Bayesian variable selection methods and demonstrate how to combine them to select genomic features associated with discrete ordinal outcomes. We compared the Bayesian and frequentist methods in terms of variable selection performance. We additionally applied the Bayesian stereotype method to an acute myeloid leukemia RNA-sequencing dataset to further demonstrate its variable selection abilities by identifying features associated with the European LeukemiaNet prognostic risk score.

When a drug is administered to exert its efficacy, it will encounter multiple barriers and go through multiple interactions. Predicting the drug-related multiple interactions is critical for drug development and safety monitoring because it provides foundations for practical, safe compatibility and rational use of multiple drugs. With the progress of artificial intelligence (AI) technology, a variety of novel prediction methods for single interaction have emerged and shown great advantages compared to the traditional, expensive and time-consuming laboratory research. To promote the comprehensive and simultaneous predictions of multiple interactions, we systematically reviewed the application of AI in drug-drug, drug-food (excipients) and drug-microbiome interactions. We began by outlining the model methods, evaluation indicators, algorithms and databases commonly used to build models for three types of drug interactions. The models based on the metabolic enzyme P450, drug similarity and drug targets have empathized among the machine learning models of drug-drug interactions. In particular, we discussed the limitations of current approaches and identified potential areas for future research. It is anticipated the in-depth review will be helpful for the development of the next-generation of systematic prediction models for simultaneous multiple interactions.

BACKGROUND: Estimation of genetic relatedness, or kinship, is used occasionally for recreational purposes and in forensic applications. While numerous methods were developed to estimate kinship, they suffer from high computational requirements and often make an untenable assumption of homogeneous population ancestry of the samples. Moreover, genetic privacy is generally overlooked in the usage of kinship estimation methods. There can be ethical concerns about finding unknown familial relationships in third-party databases. Similar ethical concerns may arise while estimating and reporting sensitive population-level statistics such as inbreeding coefficients for the concerns around marginalization and stigmatization. RESULTS: Here, we present SIGFRIED, which makes use of existing reference panels with a projection-based approach that simplifies kinship estimation in the admixed populations. We use simulated and real datasets to demonstrate the accuracy and efficiency of kinship estimation. We present a secure federated kinship estimation framework and implement a secure kinship estimator using homomorphic encryption-based primitives for computing relatedness between samples in two different sites while genotype data are kept confidential. Source code and documentation for our methods can be found at https://doi.org/10.5281/zenodo.7053352. CONCLUSIONS: Analysis of relatedness is fundamentally important for identifying relatives, in association studies, and for estimation of population-level estimates of inbreeding. As the awareness of individual and group genomic privacy is growing, privacy-preserving methods for the estimation of relatedness are needed. Presented methods alleviate the ethical and privacy concerns in the analysis of relatedness in admixed, historically isolated and underrepresented populations. SHORT ABSTRACT: Genetic relatedness is a central quantity used for finding relatives in databases, correcting biases in genome wide association studies and for estimating population-level statistics. Methods for estimating genetic relatedness have high computational requirements, and occasionally do not consider individuals from admixed ancestries. Furthermore, the ethical concerns around using genetic data and calculating relatedness are not considered. We present a projection-based approach that can efficiently and accurately estimate kinship. We implement our method using encryption-based techniques that provide provable security guarantees to protect genetic data while kinship statistics are computed among multiple sites.

To understand how distinct memories are formed and stored in the brain is an important and fundamental question in neuroscience and computational biology. A population of neurons, termed engram cells, represents the physiological manifestation of a specific memory trace and is characterized by dynamic changes in gene expression, which in turn alters the synaptic connectivity and excitability of these cells. Recent applications of single-cell RNA sequencing (scRNA-seq) and single-nucleus RNA sequencing (snRNA-seq) are promising approaches for delineating the dynamic expression profiles in these subsets of neurons, and thus understanding memory-specific genes, their combinatorial patterns and regulatory networks. The aim of this article is to review and discuss the experimental and computational procedures of sc/snRNA-seq, new studies of molecular mechanisms of memory aided by sc/snRNA-seq in human brain diseases and related mouse models, and computational challenges in understanding the regulatory mechanisms underlying long-term memory formation.

Mendelian randomization is a versatile tool to identify the possible causal relationship between an omics biomarker and disease outcome using genetic variants as instrumental variables. A key theme is the prioritization of genes whose omics readouts can be used as predictors of the disease outcome through analyzing GWAS and QTL summary data. However, there is a dearth of study of the best practice in probing the effects of multiple -omics biomarkers annotated to the same gene of interest. To bridge this gap, we propose powerful combination tests that integrate multiple correlated $P$-values without assuming the dependence structure between the exposures. Our extensive simulation experiments demonstrate the superiority of our proposed approach compared with existing methods that are adapted to the setting of our interest. The top hits of the analyses of multi-omics Alzheimer's disease datasets include genes ABCA7 and ATP1B1.

Currently, there exist no generally accepted strategies of evaluating computational models for microRNA-disease associations (MDAs). Though K-fold cross validations and case studies seem to be must-have procedures, the value of K, the evaluation metrics, and the choice of query diseases as well as the inclusion of other procedures (such as parameter sensitivity tests, ablation studies and computational cost reports) are all determined on a case-by-case basis and depending on the researchers' choices. In the current review, we include a comprehensive analysis on how 29 state-of-the-art models for predicting MDAs were evaluated. Based on the analytical results, we recommend a feasible evaluation workflow that would suit any future model to facilitate fair and systematic assessment of predictive performance.

Nucleotide and protein sequences stored in public databases are the cornerstone of many bioinformatics analyses. The records containing these sequences are prone to a wide range of errors, including incorrect functional annotation, sequence contamination and taxonomic misclassification. One source of information that can help to detect errors are the strong interdependency between records. Novel sequences in one database draw their annotations from existing records, may generate new records in multiple other locations and will have varying degrees of similarity with existing records across a range of attributes. A network perspective of these relationships between sequence records, within and across databases, offers new opportunities to detect-or even correct-erroneous entries and more broadly to make inferences about record quality. Here, we describe this novel perspective of sequence database records as a rich network, which we call the sequence database network, and illustrate the opportunities this perspective offers for quantification of database quality and detection of spurious entries. We provide an overview of the relevant databases and describe how the interdependencies between sequence records across these databases can be exploited by network analyses. We review the process of sequence annotation and provide a classification of sources of error, highlighting propagation as a major source. We illustrate the value of a network perspective through three case studies that use network analysis to detect errors, and explore the quality and quantity of critical relationships that would inform such network analyses. This systematic description of a network perspective of sequence database records provides a novel direction to combat the proliferation of errors within these critical bioinformatics resources.

The recent biotechnological progress has allowed life scientists and physicians to access an unprecedented, massive amount of data at all levels (molecular, supramolecular, cellular and so on) of biological complexity. So far, mostly classical computational efforts have been dedicated to the simulation, prediction or de novo design of biomolecules, in order to improve the understanding of their function or to develop novel therapeutics. At a higher level of complexity, the progress of omics disciplines (genomics, transcriptomics, proteomics and metabolomics) has prompted researchers to develop informatics means to describe and annotate new biomolecules identified with a resolution down to the single cell, but also with a high-throughput speed. Machine learning approaches have been implemented to both the modelling studies and the handling of biomedical data. Quantum computing (QC) approaches hold the promise to resolve, speed up or refine the analysis of a wide range of these computational problems. Here, we review and comment on recently developed QC algorithms for biocomputing, with a particular focus on multi-scale modelling and genomic analyses. Indeed, differently from other computational approaches such as protein structure prediction, these problems have been shown to be adequately mapped onto quantum architectures, the main limit for their immediate use being the number of qubits and decoherence effects in the available quantum machines. Possible advantages over the classical counterparts are highlighted, along with a description of some hybrid classical/quantum approaches, which could be the closest to be realistically applied in biocomputation.

Pan-genome analyses of metagenome-assembled genomes (MAGs) may suffer from the known issues with MAGs: fragmentation, incompleteness and contamination. Here, we conducted a critical assessment of pan-genomics of MAGs, by comparing pan-genome analysis results of complete bacterial genomes and simulated MAGs. We found that incompleteness led to significant core gene (CG) loss. The CG loss remained when using different pan-genome analysis tools (Roary, BPGA, Anvi'o) and when using a mixture of MAGs and complete genomes. Contamination had little effect on core genome size (except for Roary due to in its gene clustering issue) but had major influence on accessory genomes. Importantly, the CG loss was partially alleviated by lowering the CG threshold and using gene prediction algorithms that consider fragmented genes, but to a less degree when incompleteness was higher than 5%. The CG loss also led to incorrect pan-genome functional predictions and inaccurate phylogenetic trees. Our main findings were supported by a study of real MAG-isolate genome data. We conclude that lowering CG threshold and predicting genes in metagenome mode (as Anvi'o does with Prodigal) are necessary in pan-genome analysis of MAGs. Development of new pan-genome analysis tools specifically for MAGs are needed in future studies.

Pathway analysis has been widely used to detect pathways and functions associated with complex disease phenotypes. The proliferation of this approach is due to better interpretability of its results and its higher statistical power compared with the gene-level statistics. A plethora of pathway analysis methods that utilize multi-omics setup, rather than just transcriptomics or proteomics, have recently been developed to discover novel pathways and biomarkers. Since multi-omics gives multiple views into the same problem, different approaches are employed in aggregating these views into a comprehensive biological context. As a result, a variety of novel hypotheses regarding disease ideation and treatment targets can be formulated. In this article, we review 32 such pathway analysis methods developed for multi-omics and multi-cohort data. We discuss their availability and implementation, assumptions, supported omics types and databases, pathway analysis techniques and integration strategies. A comprehensive assessment of each method's practicality, and a thorough discussion of the strengths and drawbacks of each technique will be provided. The main objective of this survey is to provide a thorough examination of existing methods to assist potential users and researchers in selecting suitable tools for their data and analysis purposes, while highlighting outstanding challenges in the field that remain to be addressed for future development.

Well understanding protein function and structure in computational biology helps in the understanding of human beings. To face the limited proteins that are annotated structurally and functionally, the scientific community embraces the self-supervised pre-training methods from large amounts of unlabeled protein sequences for protein embedding learning. However, the protein is usually represented by individual amino acids with limited vocabulary size (e.g. 20 type proteins), without considering the strong local semantics existing in protein sequences. In this work, we propose a novel pre-training modeling approach SPRoBERTa. We first present an unsupervised protein tokenizer to learn protein representations with local fragment pattern. Then, a novel framework for deep pre-training model is introduced to learn protein embeddings. After pre-training, our method can be easily fine-tuned for different protein tasks, including amino acid-level prediction task (e.g. secondary structure prediction), amino acid pair-level prediction task (e.g. contact prediction) and also protein-level prediction task (remote homology prediction, protein function prediction). Experiments show that our approach achieves significant improvements in all tasks and outperforms the previous methods. We also provide detailed ablation studies and analysis for our protein tokenizer and training framework.

Anticancer peptides (ACPs) are bioactive peptides with antitumor activity and have become the most promising drugs in the treatment of cancer. Therefore, the accurate prediction of ACPs is of great significance to the research of cancer diseases. In the paper, we developed a more efficient prediction model called ACP_MS. Firstly, the monoMonoKGap method is used to extract the characteristic of anticancer peptide sequences and form the digital features. Then, the AdaBoost model is used to select the most discriminating features from the digital features. Finally, a stochastic gradient descent algorithm is introduced to identify anticancer peptide sequences. We adopt 7-fold cross-validation and independent test set validation, and the final accuracy of the main dataset reached 92.653% and 91.597%, respectively. The accuracy of the alternate dataset reached 98.678% and 98.317%, respectively. Compared with other advanced prediction models, the ACP_MS model improves the identification ability of anticancer peptide sequences. The data of this model can be downloaded from the public website for free https://github.com/Zhoucaimao1998/Zc.

PIWI proteins and Piwi-Interacting RNAs (piRNAs) are commonly detected in human cancers, especially in germline and somatic tissues, and correlate with poorer clinical outcomes, suggesting that they play a functional role in cancer. As the problem of combinatorial explosions between ncRNA and disease exposes gradually, new bioinformatics methods for large-scale identification and prioritization of potential associations are therefore of interest. However, in the real world, the network of interactions between molecules is enormously intricate and noisy, which poses a problem for efficient graph mining. Line graphs can extend many heterogeneous networks to replace dichotomous networks. In this study, we present a new graph neural network framework, line graph attention networks (LGAT). And we apply it to predict PiRNA disease association (GAPDA). In the experiment, GAPDA performs excellently in 5-fold cross-validation with an AUC of 0.9038. Not only that, it still has superior performance compared with methods based on collaborative filtering and attribute features. The experimental results show that GAPDA ensures the prospect of the graph neural network on such problems and can be an excellent supplement for future biomedical research.

Cell-type composition of intact bulk tissues can vary across samples. Deciphering cell-type composition and its changes during disease progression is an important step toward understanding disease pathogenesis. To infer cell-type composition, existing cell-type deconvolution methods for bulk RNA sequencing (RNA-seq) data often require matched single-cell RNA-seq (scRNA-seq) data, generated from samples with similar clinical conditions, as reference. However, due to the difficulty of obtaining scRNA-seq data in diseased samples, only limited scRNA-seq data in matched disease conditions are available. Using scRNA-seq reference to deconvolve bulk RNA-seq data from samples with different disease conditions may lead to a biased estimation of cell-type proportions. To overcome this limitation, we propose an iterative estimation procedure, MuSiC2, which is an extension of MuSiC, to perform deconvolution analysis of bulk RNA-seq data generated from samples with multiple clinical conditions where at least one condition is different from that of the scRNA-seq reference. Extensive benchmark evaluations indicated that MuSiC2 improved the accuracy of cell-type proportion estimates of bulk RNA-seq samples under different conditions as compared with the traditional MuSiC deconvolution. MuSiC2 was applied to two bulk RNA-seq datasets for deconvolution analysis, including one from human pancreatic islets and the other from human retina. We show that MuSiC2 improves current deconvolution methods and provides more accurate cell-type proportion estimates when the bulk and single-cell reference differ in clinical conditions. We believe the condition-specific cell-type composition estimates from MuSiC2 will facilitate the downstream analysis and help identify cellular targets of human diseases.

MOTIVATION: Metabolomics has developed rapidly in recent years, and metabolism-related databases are also gradually constructed. Nowadays, more and more studies are being carried out on diverse microbes, metabolites and diseases. However, the logics of various associations among microbes, metabolites and diseases are limited understanding in the biomedicine of gut microbial system. The collection and analysis of relevant microbial bioinformation play an important role in the revelation of microbe-metabolite-disease associations. Therefore, the dataset that integrates multiple relationships and the method based on complex heterogeneous graphs need to be developed. RESULTS: In this study, we integrated some databases and extracted a variety of associations data among microbes, metabolites and diseases. After obtaining the three interconnected bilateral association data (microbe-metabolite, metabolite-disease and disease-microbe), we considered building a heterogeneous graph to describe the association data. In our model, microbes were used as a bridge between diseases and metabolites. In order to fuse the information of disease-microbe-metabolite graph, we used the bipartite graph attention network on the disease-microbe and metabolite-microbe bipartite graph. The experimental results show that our model has good performance in the prediction of various disease-metabolite associations. Through the case study of type 2 diabetes mellitus, Parkinson's disease, inflammatory bowel disease and liver cirrhosis, it is noted that our proposed methodology are valuable for the mining of other associations and the prediction of biomarkers for different human diseases.Availability and implementation: https://github.com/Selenefreeze/DiMiMe.git.

In shotgun metagenomics (SM), the state-of-the-art bioinformatic workflows are referred to as high-resolution shotgun metagenomics (HRSM) and require intensive computing and disk storage resources. While the increase in data output of the latest iteration of high-throughput DNA sequencing systems can allow for unprecedented sequencing depth at a minimal cost, adjustments in HRSM workflows will be needed to properly process these ever-increasing sequence datasets. One potential adaptation is to generate so-called shallow SM datasets that contain fewer sequencing data per sample as compared with the more classic high coverage sequencing. While shallow sequencing is a promising avenue for SM data analysis, detailed benchmarks using real-data are lacking. In this case study, we took four public SM datasets, one massive and the others moderate in size and subsampled each dataset at various levels to mimic shallow sequencing datasets of various sequencing depths. Our results suggest that shallow SM sequencing is a viable avenue to obtain sound results regarding microbial community structures and that high-depth sequencing does not bring additional elements for ecological interpretation. More specifically, results obtained by subsampling as little as 0.5 M sequencing clusters per sample were similar to the results obtained with the largest subsampled dataset for human gut and agricultural soil datasets. For an Antarctic dataset, which contained only a few samples, 4 M sequencing clusters per sample was found to generate comparable results to the full dataset. One area where ultra-deep sequencing and maximizing the usage of all data was undeniably beneficial was in the generation of metagenome-assembled genomes.

Drug discovery and development is a complex and costly process. Machine learning approaches are being investigated to help improve the effectiveness and speed of multiple stages of the drug discovery pipeline. Of these, those that use Knowledge Graphs (KG) have promise in many tasks, including drug repurposing, drug toxicity prediction and target gene-disease prioritization. In a drug discovery KG, crucial elements including genes, diseases and drugs are represented as entities, while relationships between them indicate an interaction. However, to construct high-quality KGs, suitable data are required. In this review, we detail publicly available sources suitable for use in constructing drug discovery focused KGs. We aim to help guide machine learning and KG practitioners who are interested in applying new techniques to the drug discovery field, but who may be unfamiliar with the relevant data sources. The datasets are selected via strict criteria, categorized according to the primary type of information contained within and are considered based upon what information could be extracted to build a KG. We then present a comparative analysis of existing public drug discovery KGs and an evaluation of selected motivating case studies from the literature. Additionally, we raise numerous and unique challenges and issues associated with the domain and its datasets, while also highlighting key future research directions. We hope this review will motivate KGs use in solving key and emerging questions in the drug discovery domain.

In recent years, artificial intelligence (AI)/machine learning has emerged as a plausible alternative to systems biology for the elucidation of biological phenomena and in attaining specified design objective in synthetic biology. Although considered highly disruptive with numerous notable successes so far, we seek to bring attention to both the fundamental and practical pitfalls of their usage, especially in illuminating emergent behaviors from chaotic or stochastic systems in biology. Without deliberating on their suitability and the required data qualities and pre-processing approaches beforehand, the research and development community could experience similar 'AI winters' that had plagued other fields. Instead, we anticipate the integration or combination of the two approaches, where appropriate, moving forward.

MicroRNAs (miRNAs) are gene regulators involved in the pathogenesis of complex diseases such as cancers, and thus serve as potential diagnostic markers and therapeutic targets. The prerequisite for designing effective miRNA therapies is accurate discovery of miRNA-disease associations (MDAs), which has attracted substantial research interests during the last 15 years, as reflected by more than 55 000 related entries available on PubMed. Abundant experimental data gathered from the wealth of literature could effectively support the development of computational models for predicting novel associations. In 2017, Chen et al. published the first-ever comprehensive review on MDA prediction, presenting various relevant databases, 20 representative computational models, and suggestions for building more powerful ones. In the current review, as the continuation of the previous study, we revisit miRNA biogenesis, detection techniques and functions; summarize recent experimental findings related to common miRNA-associated diseases; introduce recent updates of miRNA-relevant databases and novel database releases since 2017, present mainstream webservers and new webserver releases since 2017 and finally elaborate on how fusion of diverse data sources has contributed to accurate MDA prediction.

A transcriptional regulatory network (TRN) is a collection of transcription regulators with their associated downstream genes, which is highly condition-specific. Understanding how cell states can be programmed through small molecules/drugs or conditions by modulating the whole gene expression system granted us the potential to amend abnormal cells and cure diseases. Condition Orientated Regulatory Networks (CORN, https://qinlab.sysu.edu.cn/home) is a library of condition (small molecule/drug treatments and gene knockdowns)-based transcriptional regulatory sub-networks (TRSNs) that come with an online TRSN matching tool. It allows users to browse condition-associated TRSNs or match those TRSNs by inputting transcriptomic changes of interest. CORN utilizes transcriptomic changes data after specific conditional treatment in cells, and in vivo transcription factor (TF) binding data in cells, by combining TF binding information and calculations of significant expression alterations of TFs and genes after the conditional treatments, TRNs under the effect of different conditions were constructed. In short, CORN associated 1805 different types of specific conditions (small molecule/drug treatments and gene knockdowns) to 9553 TRSNs in 25 human cell lines, involving 204TFs. By linking and curating specific conditions to responsive TRNs, the scientific community can now perceive how TRNs are altered and controlled by conditions alone in an organized manner for the first time. This study demonstrated with examples that CORN can aid the understanding of molecular pathology, pharmacology and drug repositioning, and screened drugs with high potential for cancer and coronavirus disease 2019 (COVID-19) treatments.

Antiretroviral peptides are a kind of bioactive peptides that present inhibitory activity against retroviruses through various mechanisms. Among them, viral integrase inhibitory peptides (VINIPs) are a class of antiretroviral peptides that have the ability to block the action of integrase proteins, which is essential for retroviral replication. As the number of experimentally verified bioactive peptides has increased significantly, the lack of in silico machine learning approaches can effectively predict the peptides with the integrase inhibitory activity. Here, we have developed the first prediction model for identifying the novel VINIPs using the sequence characteristics, and the hybrid feature set was considered to improve the predictive ability. The performance was evaluated by 5-fold cross-validation based on the training dataset, and the result indicates the proposed model is capable of predicting the VINIPs, with a sensitivity of 85.82%, a specificity of 88.81%, an accuracy of 88.37%, a balanced accuracy of 87.32% and a Matthews correlation coefficient value of 0.64. Most importantly, the model also consistently provides effective performance in independent testing. To sum up, we propose the first computational approach for identifying and characterizing the VINIPs, which can be considered novel antiretroviral therapy agents. Ultimately, to facilitate further research and development, iDVIP, an automatic computational tool that predicts the VINIPs has been developed, which is now freely available at http://mer.hc.mmh.org.tw/iDVIP/.

Epitope residues located on viral surface proteins are of immense interest in immunology and related applications such as vaccine development, disease diagnosis and drug design. Most tools rely on sequence-based statistical comparisons, such as information entropy of residue positions in aligned columns to infer location and properties of epitope sites. To facilitate cross-structural comparisons of epitopes on viral surface proteins, a python-based extraction tool implemented with Jupyter notebook is presented (Jupytope). Given a viral antigen structure of interest, a list of known epitope sites and a reference structure, the corresponding epitope structural properties can quickly be obtained. The tool integrates biopython modules for commonly used software such as NACCESS, DSSP as well as residue depth and outputs a list of structure-derived properties such as dihedral angles, solvent accessibility, residue depth and secondary structure that can be saved in several convenient data formats. To ensure correct spatial alignment, Jupytope takes a list of given epitope sites and their corresponding reference structure and aligns them before extracting the desired properties. Examples are demonstrated for epitopes of Influenza and severe acute respiratory syndrome coronavirus 2 (SARS-CoV2) viral strains. The extracted properties assist detection of two Influenza subtypes and show potential in distinguishing between four major clades of SARS-CoV2, as compared with randomized labels. The tool will facilitate analytical and predictive works on viral epitopes through the extracted structural information. Jupytope and extracted datasets are available at https://github.com/shamimarashid/Jupytope.

MOTIVATION: The identification of drug-target interactions (DTIs) plays a vital role for in silico drug discovery, in which the drug is the chemical molecule, and the target is the protein residues in the binding pocket. Manual DTI annotation approaches remain reliable; however, it is notoriously laborious and time-consuming to test each drug-target pair exhaustively. Recently, the rapid growth of labelled DTI data has catalysed interests in high-throughput DTI prediction. Unfortunately, those methods highly rely on the manual features denoted by human, leading to errors. RESULTS: Here, we developed an end-to-end deep learning framework called CoaDTI to significantly improve the efficiency and interpretability of drug target annotation. CoaDTI incorporates the Co-attention mechanism to model the interaction information from the drug modality and protein modality. In particular, CoaDTI incorporates transformer to learn the protein representations from raw amino acid sequences, and GraphSage to extract the molecule graph features from SMILES. Furthermore, we proposed to employ the transfer learning strategy to encode protein features by pre-trained transformer to address the issue of scarce labelled data. The experimental results demonstrate that CoaDTI achieves competitive performance on three public datasets compared with state-of-the-art models. In addition, the transfer learning strategy further boosts the performance to an unprecedented level. The extended study reveals that CoaDTI can identify novel DTIs such as reactions between candidate drugs and severe acute respiratory syndrome coronavirus 2-associated proteins. The visualization of co-attention scores can illustrate the interpretability of our model for mechanistic insights. AVAILABILITY: Source code are publicly available at https://github.com/Layne-Huang/CoaDTI.

Understanding ncRNA-protein interaction is of critical importance to unveil ncRNAs' functions. Here, we propose an integrated package LION which comprises a new method for predicting ncRNA/lncRNA-protein interaction as well as a comprehensive strategy to meet the requirement of customisable prediction. Experimental results demonstrate that our method outperforms its competitors on multiple benchmark datasets. LION can also improve the performance of some widely used tools and build adaptable models for species- and tissue-specific prediction. We expect that LION will be a powerful and efficient tool for the prediction and analysis of ncRNA/lncRNA-protein interaction. The R Package LION is available on GitHub at https://github.com/HAN-Siyu/LION/.

Extrachromosomal circular DNA (eccDNA) of chromosomal origin is found in many eukaryotic species and cell types, including cancer, where eccDNAs with oncogenes drive tumorigenesis. Most studies of eccDNA employ short-read sequencing for their identification. However, short-read sequencing cannot resolve the complexity of genomic repeats, which can lead to missing eccDNA products. Long-read sequencing technologies provide an alternative to constructing complete eccDNA maps. We present a software suite, Construction-based Rolling-circle-amplification for eccDNA Sequence Identification and Location (CReSIL), to identify and characterize eccDNA from long-read sequences. CReSIL's performance in identifying eccDNA, with a minimum F1 score of 0.98, is superior to the other bioinformatic tools based on simulated data. CReSIL provides many useful features for genomic annotation, which can be used to infer eccDNA function and Circos visualization for eccDNA architecture investigation. We demonstrated CReSIL's capability in several long-read sequencing datasets, including datasets enriched for eccDNA and whole genome datasets from cells containing large eccDNA products. In conclusion, the CReSIL suite software is a versatile tool for investigating complex and simple eccDNA in eukaryotic cells.

MOTIVATION: It has been proven that only a small fraction of the neoantigens presented by major histocompatibility complex (MHC) class I molecules on the cell surface can elicit T cells. This restriction can be attributed to the binding specificity of T cell receptor (TCR) and peptide-MHC complex (pMHC). Computational prediction of T cells binding to neoantigens is a challenging and unresolved task. RESULTS: In this paper, we proposed an attention-aware contrastive learning model, ATMTCR, to infer the TCR-pMHC binding specificity. For each TCR sequence, we used a transformer encoder to transform it to latent representation, and then masked a percentage of amino acids guided by attention weights to generate its contrastive view. Compared to fully-supervised baseline model, we verified that contrastive learning-based pretraining on large-scale TCR sequences significantly improved the prediction performance of downstream tasks. Interestingly, masking a percentage of amino acids with low attention weights yielded best performance compared to other masking strategies. Comparison experiments on two independent datasets demonstrated our method achieved better performance than other existing algorithms. Moreover, we identified important amino acids and their positional preference through attention weights, which indicated the potential interpretability of our proposed model.

Existing methods for differential network analysis could only infer whether two networks of interest have differences between two groups of samples, but could not quantify and localize network differences. In this work, a novel method, permutation-based Network True Discovery Proportions (NetTDP), is proposed to quantify the number of edges (correlations) or nodes (genes) for which the co-expression networks are different. In the NetTDP method, we propose an edge-level statistic and a node-level statistic, and detect true discoveries of edges and nodes in the sense of differential co-expression network, respectively, by the permutation-based sumSome method. Furthermore, the NetTDP method could further localize the differences by inferring the TDPs for edge or gene subsets of interest, which can be selected post hoc. Our NetTDP method allows inference on data-driven modules or biology-driven gene sets, and remains valid even when these sub-networks are optimized using the same data. Experimental results on both simulation data sets and five real data sets show the effectiveness of the proposed method in inferring the quantification and localization of differential co-expression networks. The R code is available at https://github.com/LiminLi-xjtu/NetTDP.

The three-dimensional genome structure plays a key role in cellular function and gene regulation. Single-cell Hi-C (high-resolution chromosome conformation capture) technology can capture genome structure information at the cell level, which provides the opportunity to study how genome structure varies among different cell types. Recently, a few methods are well designed for single-cell Hi-C clustering. In this manuscript, we perform an in-depth benchmark study of available single-cell Hi-C data clustering methods to implement an evaluation system for multiple clustering frameworks based on both human and mouse datasets. We compare eight methods in terms of visualization and clustering performance. Performance is evaluated using four benchmark metrics including adjusted rand index, normalized mutual information, homogeneity and Fowlkes-Mallows index. Furthermore, we also evaluate the eight methods for the task of separating cells at different stages of the cell cycle based on single-cell Hi-C data.

Accurate and effective prediction of mutation-induced protein energy change remains a great challenge and of great interest in computational biology. However, high resource consumption and insufficient structural information of proteins severely limit the experimental techniques and structure-based prediction methods. Here, we design a structure-independent protocol to accurately and effectively predict the mutation-induced protein folding free energy change with only sequence, physicochemical and evolutionary features. The proposed clustered tree regression protocol is capable of effectively exploiting the inherent data patterns by integrating unsupervised feature clustering by K-means and supervised tree regression using XGBoost, and thus enabling fast and accurate protein predictions with different mutations, with an average Pearson correlation coefficient of 0.83 and an average root-mean-square error of 0.94kcal/mol. The proposed sequence-based method not only eliminates the dependence on protein structures, but also has potential applications in protein predictions with rare structural information.

Pre-trained language models have attracted increasing attention in the biomedical domain, inspired by their great success in the general natural language domain. Among the two main branches of pre-trained language models in the general language domain, i.e. BERT (and its variants) and GPT (and its variants), the first one has been extensively studied in the biomedical domain, such as BioBERT and PubMedBERT. While they have achieved great success on a variety of discriminative downstream biomedical tasks, the lack of generation ability constrains their application scope. In this paper, we propose BioGPT, a domain-specific generative Transformer language model pre-trained on large-scale biomedical literature. We evaluate BioGPT on six biomedical natural language processing tasks and demonstrate that our model outperforms previous models on most tasks. Especially, we get 44.98%, 38.42% and 40.76% F1 score on BC5CDR, KD-DTI and DDI end-to-end relation extraction tasks, respectively, and 78.2% accuracy on PubMedQA, creating a new record. Our case study on text generation further demonstrates the advantage of BioGPT on biomedical literature to generate fluent descriptions for biomedical terms.

We live in an unprecedented time in oncology. We have accumulated samples and cases in cohorts larger and more complex than ever before. New technologies are available for quantifying solid or liquid samples at the molecular level. At the same time, we are now equipped with the computational power necessary to handle this enormous amount of quantitative data. Computational models are widely used helping us to substantiate and interpret data. Under the label of systems and precision medicine, we are putting all these developments together to improve and personalize the therapy of cancer. In this review, we use melanoma as a paradigm to present the successful application of these technologies but also to discuss possible future developments in patient care linked to them. Melanoma is a paradigmatic case for disruptive improvements in therapies, with a considerable number of metastatic melanoma patients benefiting from novel therapies. Nevertheless, a large proportion of patients does not respond to therapy or suffers from adverse events. Melanoma is an ideal case study to deploy advanced technologies not only due to the medical need but also to some intrinsic features of melanoma as a disease and the skin as an organ. From the perspective of data acquisition, the skin is the ideal organ due to its accessibility and suitability for many kinds of advanced imaging techniques. We put special emphasis on the necessity of computational strategies to integrate multiple sources of quantitative data describing the tumour at different scales and levels.

Predicting RNA solvent accessibility using only primary sequence data can be regarded as sequence-based prediction work. Currently, the established studies for sequence-based RNA solvent accessibility prediction are limited due to the available number of datasets and black box prediction. To improve these issues, we first expanded the available RNA structures and then developed a sequence-based model using modified attention layers with different receptive fields to conform to the stem-loop structure of RNA chains. We measured the improvement with an extended dataset and further explored the model's interpretability by analysing the model structures, attention values and hyperparameters. Finally, we found that the developed model regarded the pieces of a sequence as templates during the training process. This work will be helpful for researchers who would like to build RNA attribute prediction models using deep learning in the future.

Large-scale metabolomics is a powerful technique that has attracted widespread attention in biomedical studies focused on identifying biomarkers and interpreting the mechanisms of complex diseases. Despite a rapid increase in the number of large-scale metabolomic studies, the analysis of metabolomic data remains a key challenge. Specifically, diverse unwanted variations and batch effects in processing many samples have a substantial impact on identifying true biological markers, and it is a daunting challenge to annotate a plethora of peaks as metabolites in untargeted mass spectrometry-based metabolomics. Therefore, the development of an out-of-the-box tool is urgently needed to realize data integration and to accurately annotate metabolites with enhanced functions. In this study, the LargeMetabo package based on R code was developed for processing and analyzing large-scale metabolomic data. This package is unique because it is capable of (1) integrating multiple analytical experiments to effectively boost the power of statistical analysis; (2) selecting the appropriate biomarker identification method by intelligent assessment for large-scale metabolic data and (3) providing metabolite annotation and enrichment analysis based on an enhanced metabolite database. The LargeMetabo package can facilitate flexibility and reproducibility in large-scale metabolomics. The package is freely available from https://github.com/LargeMetabo/LargeMetabo.

Accurate prediction of molecular properties, such as physicochemical and bioactive properties, as well as ADME/T (absorption, distribution, metabolism, excretion and toxicity) properties, remains a fundamental challenge for molecular design, especially for drug design and discovery. In this study, we advanced a novel deep learning architecture, termed FP-GNN (fingerprints and graph neural networks), which combined and simultaneously learned information from molecular graphs and fingerprints for molecular property prediction. To evaluate the FP-GNN model, we conducted experiments on 13 public datasets, an unbiased LIT-PCBA dataset and 14 phenotypic screening datasets for breast cell lines. Extensive evaluation results showed that compared to advanced deep learning and conventional machine learning algorithms, the FP-GNN algorithm achieved state-of-the-art performance on these datasets. In addition, we analyzed the influence of different molecular fingerprints, and the effects of molecular graphs and molecular fingerprints on the performance of the FP-GNN model. Analysis of the anti-noise ability and interpretation ability also indicated that FP-GNN was competitive in real-world situations. Collectively, FP-GNN algorithm can assist chemists, biologists and pharmacists in predicting and discovering better molecules with desired functions or properties.

Glioblastoma is a fast and aggressively growing tumor in the brain and spinal cord. Mutation of amino acid residues in targets proteins, which are involved in glioblastoma, alters the structure and function and may lead to disease. In this study, we collected a set of 9386 disease-causing (drivers) mutations based on the recurrence in patient samples and experimentally annotated as pathogenic and 8728 as neutral (passenger) mutations. We observed that Arg is highly preferred at the mutant sites of drivers, whereas Met and Ile showed preferences in passengers. Inspecting neighboring residues at the mutant sites revealed that the motifs YP, CP and GRH, are preferred in drivers, whereas SI, IQ and TVI are dominant in neutral. In addition, we have computed other sequence-based features such as conservation scores, Position Specific Scoring Matrices (PSSM) and physicochemical properties, and developed a machine learning-based method, GBMDriver (GlioBlastoma Multiforme Drivers), for distinguishing between driver and passenger mutations. Our method showed an accuracy and AUC of 73.59% and 0.82, respectively, on 10-fold cross-validation and 81.99% and 0.87 in a blind set of 1809 mutants. The tool is available at https://web.iitm.ac.in/bioinfo2/GBMDriver/index.html. We envisage that the present method is helpful to prioritize driver mutations in glioblastoma and assist in identifying therapeutic targets.

Virus-encoded small RNAs (vsRNA) have been reported to play an important role in viral infection. Unfortunately, there is still a lack of an effective method for vsRNA identification. Herein, we presented vsRNAfinder, a de novo method for identifying high-confidence vsRNAs from small RNA-Seq (sRNA-Seq) data based on peak calling and Poisson distribution and is publicly available at https://github.com/ZenaCai/vsRNAfinder. vsRNAfinder outperformed two widely used methods namely miRDeep2 and ShortStack in identifying viral miRNAs with a significantly improved sensitivity. It can also be used to identify sRNAs in animals and plants with similar performance to miRDeep2 and ShortStack. vsRNAfinder would greatly facilitate effective identification of vsRNAs from sRNA-Seq data.

Subcellular localization of messenger RNAs (mRNAs) plays a key role in the spatial regulation of gene activity. The functions of mRNAs have been shown to be closely linked with their localizations. As such, understanding of the subcellular localizations of mRNAs can help elucidate gene regulatory networks. Despite several computational methods that have been developed to predict mRNA localizations within cells, there is still much room for improvement in predictive performance, especially for the multiple-location prediction. In this study, we proposed a novel multi-label multi-class predictor, termed Clarion, for mRNA subcellular localization prediction. Clarion was developed based on a manually curated benchmark dataset and leveraged the weighted series method for multi-label transformation. Extensive benchmarking tests demonstrated Clarion achieved competitive predictive performance and the weighted series method plays a crucial role in securing superior performance of Clarion. In addition, the independent test results indicate that Clarion outperformed the state-of-the-art methods and can secure accuracy of 81.47, 91.29, 79.77, 92.10, 89.15, 83.74, 80.74, 79.23 and 84.74% for chromatin, cytoplasm, cytosol, exosome, membrane, nucleolus, nucleoplasm, nucleus and ribosome, respectively. The webserver and local stand-alone tool of Clarion is freely available at http://monash.bioweb.cloud.edu.au/Clarion/.

Drug repositioning (DR) is a promising strategy to discover new indicators of approved drugs with artificial intelligence techniques, thus improving traditional drug discovery and development. However, most of DR computational methods fall short of taking into account the non-Euclidean nature of biomedical network data. To overcome this problem, a deep learning framework, namely DDAGDL, is proposed to predict drug-drug associations (DDAs) by using geometric deep learning (GDL) over heterogeneous information network (HIN). Incorporating complex biological information into the topological structure of HIN, DDAGDL effectively learns the smoothed representations of drugs and diseases with an attention mechanism. Experiment results demonstrate the superior performance of DDAGDL on three real-world datasets under 10-fold cross-validation when compared with state-of-the-art DR methods in terms of several evaluation metrics. Our case studies and molecular docking experiments indicate that DDAGDL is a promising DR tool that gains new insights into exploiting the geometric prior knowledge for improved efficacy.

The emerging ligation-free three-dimensional (3D) genome mapping technologies can identify multiplex chromatin interactions with single-molecule precision. These technologies not only offer new insight into high-dimensional chromatin organization and gene regulation, but also introduce new challenges in data visualization and analysis. To overcome these challenges, we developed MCIBox, a toolkit for multi-way chromatin interaction (MCI) analysis, including a visualization tool and a platform for identifying micro-domains with clustered single-molecule chromatin complexes. MCIBox is based on various clustering algorithms integrated with dimensionality reduction methods that can display multiplex chromatin interactions at single-molecule level, allowing users to explore chromatin extrusion patterns and super-enhancers regulation modes in transcription, and to identify single-molecule chromatin complexes that are clustered into micro-domains. Furthermore, MCIBox incorporates a two-dimensional kernel density estimation algorithm to identify micro-domains boundaries automatically. These micro-domains were stratified with distinctive signatures of transcription activity and contained different cell-cycle-associated genes. Taken together, MCIBox represents an invaluable tool for the study of multiple chromatin interactions and inaugurates a previously unappreciated view of 3D genome structure.

In the entire life cycle of drug development, the side effect is one of the major failure factors. Severe side effects of drugs that go undetected until the post-marketing stage leads to around two million patient morbidities every year in the United States. Therefore, there is an urgent need for a method to predict side effects of approved drugs and new drugs. Following this need, we present a new predictor for finding side effects of drugs. Firstly, multiple similarity matrices are constructed based on the association profile feature and drug chemical structure information. Secondly, these similarity matrices are integrated by Centered Kernel Alignment-based Multiple Kernel Learning algorithm. Then, Weighted K nearest known neighbors is utilized to complement the adjacency matrix. Next, we construct Restricted Boltzmann machines (RBM) in drug space and side effect space, respectively, and apply a penalized maximum likelihood approach to train model. At last, the average decision rule was adopted to integrate predictions from RBMs. Comparison results and case studies demonstrate, with four benchmark datasets, that our method can give a more accurate and reliable prediction result.

RNA.DNA:DNA triple helix (triplex) formation is a form of RNA-DNA interaction which regulates gene expression but is difficult to study experimentally in vivo. This makes accurate computational prediction of such interactions highly important in the field of RNA research. Current predictive methods use canonical Hoogsteen base pairing rules, which whilst biophysically valid, may not reflect the plastic nature of cell biology. Here, we present the first optimization approach to learn a probabilistic model describing RNA-DNA interactions directly from motifs derived from triplex sequencing data. We find that there are several stable interaction codes, including Hoogsteen base pairing and novel RNA-DNA base pairings, which agree with in vitro measurements. We implemented these findings in TriplexAligner, a program that uses the determined interaction codes to predict triplex binding. TriplexAligner predicts RNA-DNA interactions identified in all-to-all sequencing data more accurately than all previously published tools in human and mouse and also predicts previously studied triplex interactions with known regulatory functions. We further validated a novel triplex interaction using biophysical experiments. Our work is an important step towards better understanding of triplex formation and allows genome-wide analyses of RNA-DNA interactions.

Long non-coding RNA (lncRNA) plays important roles in a series of biological processes. The transcription of lncRNA is regulated by its promoter. Hence, accurate identification of lncRNA promoter will be helpful to understand its regulatory mechanisms. Since experimental techniques remain time consuming for gnome-wide promoter identification, developing computational tools to identify promoters are necessary. However, only few computational methods have been proposed for lncRNA promoter prediction and their performances still have room to be improved. In the present work, a convolutional neural network based model, called DeepLncPro, was proposed to identify lncRNA promoters in human and mouse. Comparative results demonstrated that DeepLncPro was superior to both state-of-the-art machine learning methods and existing models for identifying lncRNA promoters. Furthermore, DeepLncPro has the ability to extract and analyze transcription factor binding motifs from lncRNAs, which made it become an interpretable model. These results indicate that the DeepLncPro can server as a powerful tool for identifying lncRNA promoters. An open-source tool for DeepLncPro was provided at https://github.com/zhangtian-yang/DeepLncPro.

Identification of transcription factor binding sites (TFBSs) is essential to understanding of gene regulation. Designing computational models for accurate prediction of TFBSs is crucial because it is not feasible to experimentally assay all transcription factors (TFs) in all sequenced eukaryotic genomes. Although many methods have been proposed for the identification of TFBSs in humans, methods designed for plants are comparatively underdeveloped. Here, we present PlantBind, a method for integrated prediction and interpretation of TFBSs based on DNA sequences and DNA shape profiles. Built on an attention-based multi-label deep learning framework, PlantBind not only simultaneously predicts the potential binding sites of 315 TFs, but also identifies the motifs bound by transcription factors. During the training process, this model revealed a strong similarity among TF family members with respect to target binding sequences. Trans-species prediction performance using four Zea mays TFs demonstrated the suitability of this model for transfer learning. Overall, this study provides an effective solution for identifying plant TFBSs, which will promote greater understanding of transcriptional regulatory mechanisms in plants.

MOTIVATION: Single-cell/nuclei RNA-sequencing (scRNA-seq) technologies can simultaneously quantify gene expression in thousands of cells across the genome. However, the majority of the noncoding RNAs, such as microRNAs (miRNAs), cannot currently be profiled at the same scale. MiRNAs are a class of small noncoding RNAs and play an important role in gene regulation. MiRNAs originate from the processing of primary transcripts, known as primary-microRNAs (pri-miRNAs). The pri-miRNA transcripts, independent of their cognate miRNAs, can also function as long noncoding RNAs, code for micropeptides or even interact with DNA, acting like enhancers. Therefore, it is apparent that the significance of scRNA-seq pri-miRNA profiling expands beyond using pri-miRNA as proxies of mature miRNAs. However, there are no computational methods that allow profiling and quantification of pri-miRNAs at the single-cell-type resolution. RESULTS: We have developed a simple yet effective computational framework to profile pri-MiRNAs from single-cell RNA-sequencing datasets (PPMS). Based on user input, PPMS can profile pri-miRNAs at cell-type resolution. PPMS can be applied to both newly produced and publicly available datasets obtained via single cell or single-nuclei RNA-seq. It allows users to (i) investigate the distribution of pri-miRNAs across cell types and cell states and (ii) establish a relationship between the number of cells/reads sequenced and the detection of pri-miRNAs. Here, to demonstrate its efficacy, we have applied PPMS to publicly available scRNA-seq data generated from (i) individual chambers (ventricles and atria) of the human heart, (ii) human pluripotent stem cells during their differentiation into cardiomyocytes (the heart beating cells) and (iii) hiPSCs-derived cardiomyocytes infected with severe acute respiratory syndrome coronavirus 2.

In order to identify plant pentatricopeptide repeat (PPR) proteins, a framework of variable selection has been proposed. In fact, it is an effective feature selection strategy that focuses on the performance of classification. Random forest has been used as the classifier with certain variables automatically selected for discrimination between PPR functional and non-functional proteins. However, it is found that samples regarded as PPR functional proteins are wrongly classified in a high rate. In this paper, we plan to improve the framework in order to achieve better classification results. Modifications are made on the framework for better identifying PPR functional proteins. Instead of random forest, a hybrid ensemble classifier is built with its base classifiers derived from six different classification methods. Besides, an incremental strategy and a clustering by search in descending order are alternatively used for feature selection, which can effectively select the most representative variables for identification on PPR proteins. In addition, it can be found that different base classifiers alternately play an important role in the ensemble classifier with feature dimension increasing. The experimental results demonstrate the effectiveness of our improvements.

Precision medicine relies on molecular and systems biology methods as well as bidirectional association studies of phenotypes and (high-throughput) genomic data. However, the integrated use of such data often faces obstacles, especially in regards to data protection. An important prerequisite for research data processing is usually informed consent. But collecting consent is not always feasible, in particular when data are to be analyzed retrospectively. For phenotype data, anonymization, i.e. the altering of data in such a way that individuals cannot be identified, can provide an alternative. Several re-identification attacks have shown that this is a complex task and that simply removing directly identifying attributes such as names is usually not enough. More formal approaches are needed that use mathematical models to quantify risks and guide their reduction. Due to the complexity of these techniques, it is challenging and not advisable to implement them from scratch. Open software libraries and tools can provide a robust alternative. However, also the range of available anonymization tools is heterogeneous and obtaining an overview of their strengths and weaknesses is difficult due to the complexity of the problem space. We therefore performed a systematic review of open anonymization tools for structured phenotype data described in the literature between 1990 and 2021. Through a two-step eligibility assessment process, we selected 13 tools for an in-depth analysis. By comparing the supported anonymization techniques and further aspects, such as maturity, we derive recommendations for tools to use for anonymizing phenotype datasets with different properties.

Target discovery and identification processes are driven by the increasing amount of biomedical data. The vast numbers of unstructured texts of biomedical publications provide a rich source of knowledge for drug target discovery research and demand the development of specific algorithms or tools to facilitate finding disease genes and proteins. Text mining is a method that can automatically mine helpful information related to drug target discovery from massive biomedical literature. However, there is a substantial lag between biomedical publications and the subsequent abstraction of information extracted by text mining to databases. The knowledge graph is introduced to integrate heterogeneous biomedical data. Here, we describe e-TSN (Target significance and novelty explorer, http://www.lilab-ecust.cn/etsn/), a knowledge visualization web server integrating the largest database of associations between targets and diseases from the full scientific literature by constructing significance and novelty scoring methods based on bibliometric statistics. The platform aims to visualize target-disease knowledge graphs to assist in prioritizing candidate disease-related proteins. Approved drugs and associated bioactivities for each interested target are also provided to facilitate the visualization of drug-target relationships. In summary, e-TSN is a fast and customizable visualization resource for investigating and analyzing the intricate target-disease networks, which could help researchers understand the mechanisms underlying complex disease phenotypes and improve the drug discovery and development efficiency, especially for the unexpected outbreak of infectious disease pandemics like COVID-19.

Fentanyl and its analogues are psychoactive substances and the concern of fentanyl abuse has been existed in decades. Because the structure of fentanyl is easy to be modified, criminals may synthesize new fentanyl analogues to avoid supervision. The drug supervision is based on the structure matching to the database and too few kinds of fentanyl analogues are included in the database, so it is necessary to find out more potential fentanyl analogues and expand the sample space of fentanyl analogues. In this study, we introduced two deep generative models (SeqGAN and MolGPT) to generate potential fentanyl analogues, and a total of 11 041 valid molecules were obtained. The results showed that not only can we generate molecules with similar property distribution of original data, but the generated molecules also contain potential fentanyl analogues that are not pretty similar to any of original data. Ten molecules based on the rules of fentanyl analogues were selected for NMR, MS and IR validation. The results indicated that these molecules are all unreported fentanyl analogues. Furthermore, this study is the first to apply the deep learning to the generation of fentanyl analogues, greatly expands the exploring space of fentanyl analogues and provides help for the supervision of fentanyl.

The identification of long noncoding RNA (lncRNA)-disease associations is of great value for disease diagnosis and treatment, and it is now commonly used to predict potential lncRNA-disease associations with computational methods. However, the existing methods do not sufficiently extract key features during data processing, and the learning model parts are either less powerful or overly complex. Therefore, there is still potential to achieve better predictive performance by improving these two aspects. In this work, we propose a novel lncRNA-disease association prediction method LDAformer based on topological feature extraction and Transformer encoder. We construct the heterogeneous network by integrating the associations between lncRNAs, diseases and micro RNAs (miRNAs). Intra-class similarities and inter-class associations are presented as the lncRNA-disease-miRNA weighted adjacency matrix to unify semantics. Next, we design a topological feature extraction process to further obtain multi-hop topological pathway features latent in the adjacency matrix. Finally, to capture the interdependencies between heterogeneous pathways, a Transformer encoder based on the global self-attention mechanism is employed to predict lncRNA-disease associations. The efficient feature extraction and the intuitive and powerful learning model lead to ideal performance. The results of computational experiments on two datasets show that our method outperforms the state-of-the-art baseline methods. Additionally, case studies further indicate its capability to discover new associations accurately.

Long non-coding RNAs (lncRNAs) can disrupt the biological functions of protein-coding genes (PCGs) to cause cancer. However, the relationship between lncRNAs and PCGs remains unclear and difficult to predict. Machine learning has achieved a satisfactory performance in association prediction, but to our knowledge, it is currently less used in lncRNA-PCG association prediction. Therefore, we introduce GAE-LGA, a powerful deep learning model with graph autoencoders as components, to recognize potential lncRNA-PCG associations. GAE-LGA jointly explored lncRNA-PCG learning and cross-omics correlation learning for effective lncRNA-PCG association identification. The functional similarity and multi-omics similarity of lncRNAs and PCGs were accumulated and encoded by graph autoencoders to extract feature representations of lncRNAs and PCGs, which were subsequently used for decoding to obtain candidate lncRNA-PCG pairs. Comprehensive evaluation demonstrated that GAE-LGA can successfully capture lncRNA-PCG associations with strong robustness and outperformed other machine learning-based identification methods. Furthermore, multi-omics features were shown to improve the performance of lncRNA-PCG association identification. In conclusion, GAE-LGA can act as an efficient application for lncRNA-PCG association prediction with the following advantages: It fuses multi-omics information into the similarity network, making the feature representation more accurate; it can predict lncRNA-PCG associations for new lncRNAs and identify potential lncRNA-PCG associations with high accuracy.

The identification of cancer subtypes can help researchers understand hidden genomic mechanisms, enhance diagnostic accuracy and improve clinical treatments. With the development of high-throughput techniques, researchers can access large amounts of data from multiple sources. Because of the high dimensionality and complexity of multiomics and clinical data, research into the integration of multiomics data is needed, and developing effective tools for such purposes remains a challenge for researchers. In this work, we proposed an entirely unsupervised clustering method without harnessing any prior knowledge (MODEC). We used manifold optimization and deep-learning techniques to integrate multiomics data for the identification of cancer subtypes and the analysis of significant clinical variables. Since there is nonlinearity in the gene-level datasets, we used manifold optimization methodology to extract essential information from the original omics data to obtain a low-dimensional latent subspace. Then, MODEC uses a deep learning-based clustering module to iteratively define cluster centroids and assign cluster labels to each sample by minimizing the Kullback-Leibler divergence loss. MODEC was applied to six public cancer datasets from The Cancer Genome Atlas database and outperformed eight competing methods in terms of the accuracy and reliability of the subtyping results. MODEC was extremely competitive in the identification of survival patterns and significant clinical features, which could help doctors monitor disease progression and provide more suitable treatment strategies.

Multimorbidity generally refers to concurrent occurrence of multiple chronic conditions. These patients are inherently at high risk and often lead a poor quality of life due to delayed treatments. With the emergence of personalized medicine and stratified healthcare, there is a need to stratify patients right at the primary care setting. Here we developed multimorbidity analysis pipeline (MulMorPip), which can stratify patients into multimorbid subgroups or endotypes based on their lifetime disease diagnosis and characterize them based on demographic features and underlying disease-disease interaction networks. By implementing MulMorPip on UK Biobank cohort, we report five distinct molecular subclasses or endotypes of multimorbidity. For each patient, we calculated the existence of broad disease classes defined by Charlson's comorbidity classification using the International Classification of Diseases-10 encoding. We then applied multiple correspondence analysis in 77 524 patients from UK Biobank, who had multimorbidity of more than one disease, which resulted in five multimorbid clusters. We further validated these clusters using machine learning and were able to classify 20% model-blind test set patients with an accuracy of 97% and an average Jaccard similarity of 84%. This was followed by demographic characterization and development of interlinking disease network for each cluster to understand disease-disease interactions. Our identified five endotypes of multimorbidity draw attention to dementia, stroke and paralysis as important drivers of multimorbidity stratification. Inclusion of such patient stratification at the primary care setting can help general practitioners to better observe patients' multiple chronic conditions, their risk stratification and personalization of treatment strategies.

Recent advances in Knowledge Graphs (KGs) and Knowledge Graph Embedding Models (KGEMs) have led to their adoption in a broad range of fields and applications. The current publishing system in machine learning requires newly introduced KGEMs to achieve state-of-the-art performance, surpassing at least one benchmark in order to be published. Despite this, dozens of novel architectures are published every year, making it challenging for users, even within the field, to deduce the most suitable configuration for a given application. A typical biomedical application of KGEMs is drug-disease prediction in the context of drug discovery, in which a KGEM is trained to predict triples linking drugs and diseases. These predictions can be later tested in clinical trials following extensive experimental validation. However, given the infeasibility of evaluating each of these predictions and that only a minimal number of candidates can be experimentally tested, models that yield higher precision on the top prioritized triples are preferred. In this paper, we apply the concept of ensemble learning on KGEMs for drug discovery to assess whether combining the predictions of several models can lead to an overall improvement in predictive performance. First, we trained and benchmarked 10 KGEMs to predict drug-disease triples on two independent biomedical KGs designed for drug discovery. Following, we applied different ensemble methods that aggregate the predictions of these models by leveraging the distribution or the position of the predicted triple scores. We then demonstrate how the ensemble models can achieve better results than the original KGEMs by benchmarking the precision (i.e., number of true positives prioritized) of their top predictions. Lastly, we released the source code presented in this work at https://github.com/enveda/kgem-ensembles-in-drug-discovery.

Gene expression in mammalian cells is inherently stochastic and mRNAs are synthesized in discrete bursts. Single-cell transcriptomics provides an unprecedented opportunity to explore the transcriptome-wide kinetics of transcriptional bursting. However, current analysis methods provide limited accuracy in bursting inference due to substantial noise inherent to single-cell transcriptomic data. In this study, we developed BISC, a Bayesian method for inferring bursting parameters from single cell transcriptomic data. Based on a beta-gamma-Poisson model, BISC modeled the mean-variance dependency to achieve accurate estimation of bursting parameters from noisy data. Evaluation based on both simulation and real intron sequential RNA fluorescence in situ hybridization data showed improved accuracy and reliability of BISC over existing methods, especially for genes with low expression values. Further application of BISC found bursting frequency but not bursting size was strongly associated with gene expression regulation. Moreover, our analysis provided new mechanistic insights into the functional role of enhancer and superenhancer by modulating both bursting frequency and size. BISC also formulated a downstream framework to identify differential bursting (in frequency and size separately) genes in samples under different conditions. Applying to multiple datasets (a mouse embryonic cell and fibroblast dataset, a human immune cell dataset and a human pancreatic cell dataset), BISC identified known cell-type signature genes that were missed by differential expression analysis, providing additional insights in understanding the cell-specific stochastic gene transcription. Applying to datasets of human lung and colon cancers, BISC successfully detected tumor signature genes based on alterations in bursting kinetics, which illustrates its value in understanding disease development regarding transcriptional bursting. Collectively, BISC provides a new tool for accurately inferring bursting kinetics and detecting differential bursting genes. This study also produced new insights in the role of transcriptional bursting in regulating gene expression, cell identity and tumor progression.

The identification of disease-causing genes is critical for mechanistic understanding of disease etiology and clinical manipulation in disease prevention and treatment. Yet the existing approaches in tackling this question are inadequate in accuracy and efficiency, demanding computational methods with higher identification power. Here, we proposed a new method called DGHNE to identify disease-causing genes through a heterogeneous biomedical network empowered by network enhancement. First, a disease-disease association network was constructed by the cosine similarity scores between phenotype annotation vectors of diseases, and a new heterogeneous biomedical network was constructed by using disease-gene associations to connect the disease-disease network and gene-gene network. Then, the heterogeneous biomedical network was further enhanced by using network embedding based on the Gaussian random projection. Finally, network propagation was used to identify candidate genes in the enhanced network. We applied DGHNE together with five other methods into the most updated disease-gene association database termed DisGeNet. Compared with all other methods, DGHNE displayed the highest area under the receiver operating characteristic curve and the precision-recall curve, as well as the highest precision and recall, in both the global 5-fold cross-validation and predicting new disease-gene associations. We further performed DGHNE in identifying the candidate causal genes of Parkinson's disease and diabetes mellitus, and the genes connecting hyperglycemia and diabetes mellitus. In all cases, the predicted causing genes were enriched in disease-associated gene ontology terms and Kyoto Encyclopedia of Genes and Genomes pathways, and the gene-disease associations were highly evidenced by independent experimental studies.

Emerging evidence suggests that circular RNA (circRNA) is an important regulator of a variety of pathological processes and serves as a promising biomarker for many complex human diseases. Nevertheless, there are relatively few known circRNA-disease associations, and uncovering new circRNA-disease associations by wet-lab methods is time consuming and costly. Considering the limitations of existing computational methods, we propose a novel approach named MNMDCDA, which combines high-order graph convolutional networks (high-order GCNs) and deep neural networks to infer associations between circRNAs and diseases. Firstly, we computed different biological attribute information of circRNA and disease separately and used them to construct multiple multi-source similarity networks. Then, we used the high-order GCN algorithm to learn feature embedding representations with high-order mixed neighborhood information of circRNA and disease from the constructed multi-source similarity networks, respectively. Finally, the deep neural network classifier was implemented to predict associations of circRNAs with diseases. The MNMDCDA model obtained AUC scores of 95.16%, 94.53%, 89.80% and 91.83% on four benchmark datasets, i.e., CircR2Disease, CircAtlas v2.0, Circ2Disease and CircRNADisease, respectively, using the 5-fold cross-validation approach. Furthermore, 25 of the top 30 circRNA-disease pairs with the best scores of MNMDCDA in the case study were validated by recent literature. Numerous experimental results indicate that MNMDCDA can be used as an effective computational tool to predict circRNA-disease associations and can provide the most promising candidates for biological experiments.

The rapid development of biomedicine has produced a large number of biomedical written materials. These unstructured text data create serious challenges for biomedical researchers to find information. Biomedical named entity recognition (BioNER) and biomedical relation extraction (BioRE) are the two most fundamental tasks of biomedical text mining. Accurately and efficiently identifying entities and extracting relations have become very important. Methods that perform two tasks separately are called pipeline models, and they have shortcomings such as insufficient interaction, low extraction quality and easy redundancy. To overcome the above shortcomings, many deep learning-based joint name entity recognition and relation extraction models have been proposed, and they have achieved advanced performance. This paper comprehensively summarize deep learning models for joint name entity recognition and relation extraction for biomedicine. The joint BioNER and BioRE models are discussed in the light of the challenges existing in the BioNER and BioRE tasks. Five joint BioNER and BioRE models and one pipeline model are selected for comparative experiments on four biomedical public datasets, and the experimental results are analyzed. Finally, we discuss the opportunities for future development of deep learning-based joint BioNER and BioRE models.

The ongoing coronavirus disease 2019 (COVID-19) pandemic has highlighted the need to better understand virus-host interactions. We developed a network-based method that expands the severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2)-host protein interaction network and identifies host targets that modulate viral infection. To disrupt the SARS-CoV-2 interactome, we systematically probed for potent compounds that selectively target the identified host proteins with high expression in cells relevant to COVID-19. We experimentally tested seven chemical inhibitors of the identified host proteins for modulation of SARS-CoV-2 infection in human cells that express ACE2 and TMPRSS2. Inhibition of the epigenetic regulators bromodomain-containing protein 4 (BRD4) and histone deacetylase 2 (HDAC2), along with ubiquitin-specific peptidase (USP10), enhanced SARS-CoV-2 infection. Such proviral effect was observed upon treatment with compounds JQ1, vorinostat, romidepsin and spautin-1, when measured by cytopathic effect and validated by viral RNA assays, suggesting that the host proteins HDAC2, BRD4 and USP10 have antiviral functions. We observed marked differences in antiviral effects across cell lines, which may have consequences for identification of selective modulators of viral infection or potential antiviral therapeutics. While network-based approaches enable systematic identification of host targets and selective compounds that may modulate the SARS-CoV-2 interactome, further developments are warranted to increase their accuracy and cell-context specificity.

Short hairpin RNA (shRNA)-mediated gene silencing is an important technology to achieve RNA interference, in which the design of potent and reliable shRNA molecules plays a crucial role. However, efficient shRNA target selection through biological technology is expensive and time consuming. Hence, it is crucial to develop a more precise and efficient computational method to design potent and reliable shRNA molecules. In this work, we present an interpretable classification model for the shRNA target prediction using the Light Gradient Boosting Machine algorithm called ILGBMSH. Rather than utilizing only the shRNA sequence feature, we extracted 554 biological and deep learning features, which were not considered in previous shRNA prediction research. We evaluated the performance of our model compared with the state-of-the-art shRNA target prediction models. Besides, we investigated the feature explanation from the model's parameters and interpretable method called Shapley Additive Explanations, which provided us with biological insights from the model. We used independent shRNA experiment data from other resources to prove the predictive ability and robustness of our model. Finally, we used our model to design the miR30-shRNA sequences and conducted a gene knockdown experiment. The experimental result was perfectly in correspondence with our expectation with a Pearson's coefficient correlation of 0.985. In summary, the ILGBMSH model can achieve state-of-the-art shRNA prediction performance and give biological insights from the machine learning model parameters.

MOTIVATION: Discovering the drug-target interactions (DTIs) is a crucial step in drug development such as the identification of drug side effects and drug repositioning. Since identifying DTIs by web-biological experiments is time-consuming and costly, many computational-based approaches have been proposed and have become an efficient manner to infer the potential interactions. Although extensive effort is invested to solve this task, the prediction accuracy still needs to be improved. More especially, heterogeneous network-based approaches do not fully consider the complex structure and rich semantic information in these heterogeneous networks. Therefore, it is still a challenge to predict DTIs efficiently. RESULTS: In this study, we develop a novel method via Multiview heterogeneous information network embedding with Hierarchical Attention mechanisms to discover potential Drug-Target Interactions (MHADTI). Firstly, MHADTI constructs different similarity networks for drugs and targets by utilizing their multisource information. Combined with the known DTI network, three drug-target heterogeneous information networks (HINs) with different views are established. Secondly, MHADTI learns embeddings of drugs and targets from multiview HINs with hierarchical attention mechanisms, which include the node-level, semantic-level and graph-level attentions. Lastly, MHADTI employs the multilayer perceptron to predict DTIs with the learned deep feature representations. The hierarchical attention mechanisms could fully consider the importance of nodes, meta-paths and graphs in learning the feature representations of drugs and targets, which makes their embeddings more comprehensively. Extensive experimental results demonstrate that MHADTI performs better than other SOTA prediction models. Moreover, analysis of prediction results for some interested drugs and targets further indicates that MHADTI has advantages in discovering DTIs. AVAILABILITY AND IMPLEMENTATION: https://github.com/pxystudy/MHADTI.

Antimicrobial peptides (AMPs) have received a great deal of attention given their potential to become a plausible option to fight multi-drug resistant bacteria as well as other pathogens. Quantitative sequence-activity models (QSAMs) have been helpful to discover new AMPs because they allow to explore a large universe of peptide sequences and help reduce the number of wet lab experiments. A main aspect in the building of QSAMs based on shallow learning is to determine an optimal set of protein descriptors (features) required to discriminate between sequences with different antimicrobial activities. These features are generally handcrafted from peptide sequence datasets that are labeled with specific antimicrobial activities. However, recent developments have shown that unsupervised approaches can be used to determine features that outperform human-engineered (handcrafted) features. Thus, knowing which of these two approaches contribute to a better classification of AMPs, it is a fundamental question in order to design more accurate models. Here, we present a systematic and rigorous study to compare both types of features. Experimental outcomes show that non-handcrafted features lead to achieve better performances than handcrafted features. However, the experiments also prove that an improvement in performance is achieved when both types of features are merged. A relevance analysis reveals that non-handcrafted features have higher information content than handcrafted features, while an interaction-based importance analysis reveals that handcrafted features are more important. These findings suggest that there is complementarity between both types of features. Comparisons regarding state-of-the-art deep models show that shallow models yield better performances both when fed with non-handcrafted features alone and when fed with non-handcrafted and handcrafted features together.

Glutarylation is a post-translational modification which plays an irreplaceable role in various functions of the cell. Therefore, it is very important to accurately identify the glutarylation substrates and its corresponding glutarylation sites. In recent years, many computational methods of glutarylation sites have emerged one after another, but there are still many limitations, among which noisy data and the class imbalance problem caused by the uncertainty of non-glutarylation sites are great challenges. In this study, we propose a new semi-supervised learning algorithm, named FCCCSR, to identify reliable non-glutarylation lysine sites from unlabeled samples as negative samples. FCCCSR first finds core objects from positive samples according to reverse nearest neighbor information, and then clusters core objects based on natural neighbor structure. Finally, reliable negative samples are selected according to clustering result. With FCCCSR algorithm, we propose a new method named FCCCSR_Glu for glutarylation sites identification. In this study, multi-view features are extracted and fused to describe peptides, including amino acid composition, BLOSUM62, amino acid factors and composition of k-spaced amino acid pairs. Then, reliable negative samples selected by FCCCSR and positive samples are combined to establish models and XGBoost optimized by differential evolution algorithm is used as the classifier. On the independent testing dataset, FCCCSR_Glu achieves 85.18%, 98.36%, 94.31% and 0.8651 in sensitivity, specificity, accuracy and Matthew's Correlation Coefficient, respectively, which is superior to state-of-the-art methods in predicting glutarylation sites. Therefore, FCCCSR_Glu can be a useful tool for glutarylation sites prediction and FCCCSR algorithm can effectively select reliable negative samples from unlabeled samples. The data and code are available on https://github.com/xbbxhbc/FCCCSR_Glu.git.

Identification of new chemical compounds with desired structural diversity and biological properties plays an essential role in drug discovery, yet the construction of such a potential space with elements of 'near-drug' properties is still a challenging task. In this work, we proposed a multimodal chemical information reconstruction system to automatically process, extract and align heterogeneous information from the text descriptions and structural images of chemical patents. Our key innovation lies in a heterogeneous data generator that produces cross-modality training data in the form of text descriptions and Markush structure images, from which a two-branch model with image- and text-processing units can then learn to both recognize heterogeneous chemical entities and simultaneously capture their correspondence. In particular, we have collected chemical structures from ChEMBL database and chemical patents from the European Patent Office and the US Patent and Trademark Office using keywords 'A61P, compound, structure' in the years from 2010 to 2020, and generated heterogeneous chemical information datasets with 210K structural images and 7818 annotated text snippets. Based on the reconstructed results and substituent replacement rules, structural libraries of a huge number of near-drug compounds can be generated automatically. In quantitative evaluations, our model can correctly reconstruct 97% of the molecular images into structured format and achieve an F1-score around 97-98% in the recognition of chemical entities, which demonstrated the effectiveness of our model in automatic information extraction from chemical patents, and hopefully transforming them to a user-friendly, structured molecular database enriching the near-drug space to realize the intelligent retrieval technology of chemical knowledge.

MicroRNAs (miRNAs) are closely related to a variety of human diseases, not only regulating gene expression, but also having an important role in human life activities and being viable targets of small molecule drugs for disease treatment. Current computational techniques to predict the potential associations between small molecule and miRNA are not that accurate. Here, we proposed a new computational method based on a deep autoencoder and a scalable tree boosting model (DAESTB), to predict associations between small molecule and miRNA. First, we constructed a high-dimensional feature matrix by integrating small molecule-small molecule similarity, miRNA-miRNA similarity and known small molecule-miRNA associations. Second, we reduced feature dimensionality on the integrated matrix using a deep autoencoder to obtain the potential feature representation of each small molecule-miRNA pair. Finally, a scalable tree boosting model is used to predict small molecule and miRNA potential associations. The experiments on two datasets demonstrated the superiority of DAESTB over various state-of-the-art methods. DAESTB achieved the best AUC value. Furthermore, in three case studies, a large number of predicted associations by DAESTB are confirmed with the public accessed literature. We envision that DAESTB could serve as a useful biological model for predicting potential small molecule-miRNA associations.

Identifying synergistic drug combinations (SDCs) is a great challenge due to the combinatorial complexity and the fact that SDC is cell line specific. The existing computational methods either did not consider the cell line specificity of SDC, or did not perform well by building model for each cell line independently. In this paper, we present a novel encoder-decoder network named SDCNet for predicting cell line-specific SDCs. SDCNet learns common patterns across different cell lines as well as cell line-specific features in one model for drug combinations. This is realized by considering the SDC graphs of different cell lines as a relational graph, and constructing a relational graph convolutional network (R-GCN) as the encoder to learn and fuse the deep representations of drugs for different cell lines. An attention mechanism is devised to integrate the drug features from different layers of the R-GCN according to their relative importance so that representation learning is further enhanced. The common patterns are exploited through partial parameter sharing in cell line-specific decoders, which not only reconstruct the known SDCs but also predict new ones for each cell line. Experiments on various datasets demonstrate that SDCNet is superior to state-of-the-art methods and is also robust when generalized to new cell lines that are different from the training ones. Finally, the case study again confirms the effectiveness of our method in predicting novel reliable cell line-specific SDCs.

More than one-third of the proteins contain metal ions in the Protein Data Bank. Correct identification of metal ion-binding residues is important for understanding protein functions and designing novel drugs. Due to the small size and high versatility of metal ions, it remains challenging to computationally predict their binding sites from protein sequence. Existing sequence-based methods are of low accuracy due to the lack of structural information, and time-consuming owing to the usage of multi-sequence alignment. Here, we propose LMetalSite, an alignment-free sequence-based predictor for binding sites of the four most frequently seen metal ions in BioLiP (Zn2+, Ca2+, Mg2+ and Mn2+). LMetalSite leverages the pretrained language model to rapidly generate informative sequence representations and employs transformer to capture long-range dependencies. Multi-task learning is adopted to compensate for the scarcity of training data and capture the intrinsic similarities between different metal ions. LMetalSite was shown to surpass state-of-the-art structure-based methods by more than 19.7, 14.4, 36.8 and 12.6% in area under the precision recall on the four independent tests, respectively. Further analyses indicated that the self-attention modules are effective to learn the structural contexts of residues from protein sequence. We provide the data sets, source codes and trained models of LMetalSite at https://github.com/biomed-AI/LMetalSite.

Metagenomic sequencing analysis (mNGS) has been implemented as an alternative approach for pathogen diagnosis in recent years, which is independent of cultivation and is able to identify all potential antibiotic resistance genes (ARGs). However, current mNGS methods have to deal with low amounts of prokaryotic deoxyribonucleic acid (DNA) and high amounts of host DNA in clinical samples, which significantly decrease the overall microbial detection resolution. The recently released nanopore adaptive sampling (NAS) technology facilitates immediate mapping of individual nucleotides to a given reference as each molecule is sequenced. User-defined thresholds allow for the retention or rejection of specific molecules, informed by the real-time reference mapping results, as they are physically passing through a given sequencing nanopore. We developed a metagenomics workflow for ultra-sensitive diagnosis of bacterial pathogens and ARGs from clinical samples, which is based on the efficient selective 'human host depletion' NAS sequencing, real-time species identification and species-specific resistance gene prediction. Our method increased the microbial sequence yield at least 8-fold in all 21 sequenced clinical Bronchoalveolar Lavage Fluid (BALF) samples (4.5 h from sample to result) and accurately detected the ARGs at species level. The species-level positive percent agreement between metagenomic sequencing and laboratory culturing was 100% (16/16) and negative percent agreement was 100% (5/5) in our approach. Further work is required for a more robust validation of our approach with large sample size to allow its application to other infection types.

Long non-coding RNA (lncRNA) and microRNA (miRNA) are two typical types of non-coding RNAs (ncRNAs), their interaction plays an important regulatory role in many biological processes. Exploring the interactions between unknown lncRNA and miRNA can help us better understand the functional expression between lncRNA and miRNA. At present, the interactions between lncRNA and miRNA are mainly obtained through biological experiments, but such experiments are often time-consuming and labor-intensive, it is necessary to design a computational method that can predict the interactions between lncRNA and miRNA. In this paper, we propose a method based on graph convolutional neural (GCN) network and conditional random field (CRF) for predicting human lncRNA-miRNA interactions, named GCNCRF. First, we construct a heterogeneous network using the known interactions of lncRNA and miRNA in the LncRNASNP2 database, the lncRNA/miRNA integration similarity network, and the lncRNA/miRNA feature matrix. Second, the initial embedding of nodes is obtained using a GCN network. A CRF set in the GCN hidden layer can update the obtained preliminary embeddings so that similar nodes have similar embeddings. At the same time, an attention mechanism is added to the CRF layer to reassign weights to nodes to better grasp the feature information of important nodes and ignore some nodes with less influence. Finally, the final embedding is decoded and scored through the decoding layer. Through a 5-fold cross-validation experiment, GCNCRF has an area under the receiver operating characteristic curve value of 0.947 on the main dataset, which has higher prediction accuracy than the other six state-of-the-art methods.

Exiting computational models for drug-target binding affinity prediction have much room for improvement in prediction accuracy, robustness and generalization ability. Most deep learning models lack interpretability analysis and few studies provide application examples. Based on these observations, we presented a novel model named Molecule Representation Block-based Drug-Target binding Affinity prediction (MRBDTA). MRBDTA is composed of embedding and positional encoding, molecule representation block and interaction learning module. The advantages of MRBDTA are reflected in three aspects: (i) developing Trans block to extract molecule features through improving the encoder of transformer, (ii) introducing skip connection at encoder level in Trans block and (iii) enhancing the ability to capture interaction sites between proteins and drugs. The test results on two benchmark datasets manifest that MRBDTA achieves the best performance compared with 11 state-of-the-art models. Besides, through replacing Trans block with single Trans encoder and removing skip connection in Trans block, we verified that Trans block and skip connection could effectively improve the prediction accuracy and reliability of MRBDTA. Then, relying on multi-head attention mechanism, we performed interpretability analysis to illustrate that MRBDTA can correctly capture part of interaction sites between proteins and drugs. In case studies, we firstly employed MRBDTA to predict binding affinities between Food and Drug Administration-approved drugs and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) replication-related proteins. Secondly, we compared true binding affinities between 3C-like proteinase and 185 drugs with those predicted by MRBDTA. The final results of case studies reveal reliable performance of MRBDTA in drug design for SARS-CoV-2.

The discovery and repurposing of drugs require a deep understanding of the mechanism of drug action (MODA). Existing computational methods mainly model MODA with the protein-protein interaction (PPI) network. However, the molecular interactions of drugs in the human body are far beyond PPIs. Additionally, the lack of interpretability of these models hinders their practicability. We propose an interpretable deep learning-based path-reasoning framework (iDPath) for drug discovery and repurposing by capturing MODA on by far the most comprehensive multilayer biological network consisting of the complex high-dimensional molecular interactions between genes, proteins and chemicals. Experiments show that iDPath outperforms state-of-the-art machine learning methods on a general drug repurposing task. Further investigations demonstrate that iDPath can identify explicit critical paths that are consistent with clinical evidence. To demonstrate the practical value of iDPath, we apply it to the identification of potential drugs for treating prostate cancer and hypertension. Results show that iDPath can discover new FDA-approved drugs. This research provides a novel interpretable artificial intelligence perspective on drug discovery.

In recent years, many studies have illustrated the significant role that non-coding RNA (ncRNA) plays in biological activities, in which lncRNA, miRNA and especially their interactions have been proved to affect many biological processes. Some in silico methods have been proposed and applied to identify novel lncRNA-miRNA interactions (LMIs), but there are still imperfections in their RNA representation and information extraction approaches, which imply there is still room for further improving their performances. Meanwhile, only a few of them are accessible at present, which limits their practical applications. The construction of a new tool for LMI prediction is thus imperative for the better understanding of their relevant biological mechanisms. This study proposed a novel method, ncRNAInter, for LMI prediction. A comprehensive strategy for RNA representation and an optimized deep learning algorithm of graph neural network were utilized in this study. ncRNAInter was robust and showed better performance of 26.7% higher Matthews correlation coefficient than existing reputable methods for human LMI prediction. In addition, ncRNAInter proved its universal applicability in dealing with LMIs from various species and successfully identified novel LMIs associated with various diseases, which further verified its effectiveness and usability. All source code and datasets are freely available at https://github.com/idrblab/ncRNAInter.

Imaging genetics provides unique insights into the pathological studies of complex brain diseases by integrating the characteristics of multi-level medical data. However, most current imaging genetics research performs incomplete data fusion. Also, there is a lack of effective deep learning methods to analyze neuroimaging and genetic data jointly. Therefore, this paper first constructs the brain region-gene networks to intuitively represent the association pattern of pathogenetic factors. Second, a novel feature information aggregation model is constructed to accurately describe the information aggregation process among brain region nodes and gene nodes. Finally, a deep learning method called feature information aggregation and diffusion generative adversarial network (FIAD-GAN) is proposed to efficiently classify samples and select features. We focus on improving the generator with the proposed convolution and deconvolution operations, with which the interpretability of the deep learning framework has been dramatically improved. The experimental results indicate that FIAD-GAN can not only achieve superior results in various disease classification tasks but also extract brain regions and genes closely related to AD. This work provides a novel method for intelligent clinical decisions. The relevant biomedical discoveries provide a reliable reference and technical basis for the clinical diagnosis, treatment and pathological analysis of disease.

Short open reading frames (sORFs) refer to the small nucleic fragments no longer than 303 nt in length that probably encode small peptides. To date, translatable sORFs have been found in both untranslated regions of messenger ribonucleic acids (RNAs; mRNAs) and long non-coding RNAs (lncRNAs), playing vital roles in a myriad of biological processes. As not all sORFs are translated or essentially translatable, it is important to develop a highly accurate computational tool for characterizing the coding potential of sORFs, thereby facilitating discovery of novel functional peptides. In light of this, we designed a series of ensemble models by integrating Efficient-CapsNet and LightGBM, collectively termed csORF-finder, to differentiate the coding sORFs (csORFs) from non-coding sORFs in Homo sapiens, Mus musculus and Drosophila melanogaster, respectively. To improve the performance of csORF-finder, we introduced a novel feature encoding scheme named trinucleotide deviation from expected mean (TDE) and computed all types of in-frame sequence-based features, such as i-framed-3mer, i-framed-CKSNAP and i-framed-TDE. Benchmarking results showed that these features could significantly boost the performance compared to the original 3-mer, CKSNAP and TDE features. Our performance comparisons showed that csORF-finder achieved a superior performance than the state-of-the-art methods for csORF prediction on multi-species and non-ATG initiation independent test datasets. Furthermore, we applied csORF-finder to screen the lncRNA datasets for identifying potential csORFs. The resulting data serve as an important computational repository for further experimental validation. We hope that csORF-finder can be exploited as a powerful platform for high-throughput identification of csORFs and functional characterization of these csORFs encoded peptides.

Robust strategies to identify patients at high risk for tumor metastasis, such as those frequently observed in intrahepatic cholangiocarcinoma (ICC), remain limited. While gene/protein expression profiling holds great potential as an approach to cancer diagnosis and prognosis, previously developed protocols using multiple diagnostic signatures for expression-based metastasis prediction have not been widely applied successfully because batch effects and different data types greatly decreased the predictive performance of gene/protein expression profile-based signatures in interlaboratory and data type dependent validation. To address this problem and assist in more precise diagnosis, we performed a genome-wide integrative proteome and transcriptome analysis and developed an ensemble machine learning-based integration algorithm for metastasis prediction (EMLI-Metastasis) and risk stratification (EMLI-Prognosis) in ICC. Based on massive proteome (216) and transcriptome (244) data sets, 132 feature (biomarker) genes were selected and used to train the EMLI-Metastasis algorithm. To accurately detect the metastasis of ICC patients, we developed a weighted ensemble machine learning method based on k-Top Scoring Pairs (k-TSP) method. This approach generates a metastasis classifier for each bootstrap aggregating training data set. Ten binary expression rank-based classifiers were generated for detection of metastasis separately. To further improve the accuracy of the method, the 10 binary metastasis classifiers were combined by weighted voting based on the score from the prediction results of each classifier. The prediction accuracy of the EMLI-Metastasis algorithm achieved 97.1% and 85.0% in proteome and transcriptome datasets, respectively. Among the 132 feature genes, 21 gene-pair signatures were developed to establish a metastasis-related prognosis risk-stratification model in ICC (EMLI-Prognosis). Based on EMLI-Prognosis algorithm, patients in the high-risk group had significantly dismal overall survival relative to the low-risk group in the clinical cohort (P-value < 0.05). Taken together, the EMLI-ICC algorithm provides a powerful and robust means for accurate metastasis prediction and risk stratification across proteome and transcriptome data types that is superior to currently used clinicopathological features in patients with ICC. Our developed algorithm could have profound implications not just in improved clinical care in cancer metastasis risk prediction, but also more broadly in machine-learning-based multi-cohort diagnosis method development. To make the EMLI-ICC algorithm easily accessible for clinical application, we established a web-based server for metastasis risk prediction (http://ibi.zju.edu.cn/EMLI/).

Sarcopenia is correlated with poor clinical outcomes in breast cancer (BC) patients. However, there is no precise quantitative study on the correlation between body composition changes and BC metastasis and survival. The present study proposed a deep learning radiomics (DLR) approach to investigate the effects of muscle and fat on distant metastasis and death outcomes in BC patients. Image feature extraction was performed on 4th thoracic vertebra (T4) and 11th thoracic vertebra (T11) on computed tomography (CT) image levels by DLR, and image features were combined with clinical information to predict distant metastasis in BC patients. Clinical information combined with DLR significantly predicted distant metastasis in BC patients. In the test cohort, the area under the curve of model performance on clinical information combined with DLR was 0.960 (95% CI: 0.942-0.979, P < 0.001). The patients with distant metastases had a lower pectoral muscle index in T4 (PMI/T4) than in patients without metastases. PMI/T4 and visceral fat tissue area in T11 (VFA/T11) were independent prognostic factors for the overall survival in BC patients. The pectoralis muscle area in T4 (PMA/T4) and PMI/T4 is an independent prognostic factor for distant metastasis-free survival in BC patients. The current study further confirmed that muscle/fat of T4 and T11 levels have a significant effect on the distant metastasis of BC. Appending the network features of T4 and T11 to the model significantly enhances the prediction performance of distant metastasis of BC, providing a valuable biomarker for the early treatment of BC patients.

Most polygenic risk score (PRS)models have been based on data from populations of European origins (accounting for the majority of the large genomics datasets, e.g. >78% in the UK Biobank and >85% in the GTEx project). Although several large-scale Asian biobanks were initiated (e.g. Japanese, Korean, Han Chinese biobanks), most other Asian countries have little or near-zero genomics data. To implement PRS models for under-represented populations, we explored transfer learning approaches, assuming that information from existing large datasets can compensate for the small sample size that can be feasibly obtained in developing countries, like Vietnam. Here, we benchmark 13 common PRS methods in meta-population strategy (combining individual genotype data from multiple populations) and multi-population strategy (combining summary statistics from multiple populations). Our results highlight the complementarity of different populations and the choice of methods should depend on the target population. Based on these results, we discussed a set of guidelines to help users select the best method for their datasets. We developed a robust and comprehensive software to allow for benchmarking comparisons between methods and proposed a computational framework for improving PRS performance in a dataset with a small sample size. This work is expected to inform the development of genomics applications in under-represented populations. PRSUP framework is available at: https://github.com/BiomedicalMachineLearning/VGP.

Time-course single-cell RNA sequencing (scRNA-seq) data have been widely used to explore dynamic changes in gene expression of transcription factors (TFs) and their target genes. This information is useful to reconstruct cell-type-specific gene regulatory networks (GRNs). However, the existing tools are commonly designed to analyze either time-course bulk gene expression data or static scRNA-seq data via pseudo-time cell ordering. A few methods successfully utilize the information from multiple time points while also considering the characteristics of scRNA-seq data. We proposed dynDeepDRIM, a novel deep learning model to reconstruct GRNs using time-course scRNA-seq data. It represents the joint expression of a gene pair as an image and utilizes the image of the target TF-gene pair and the ones of the potential neighbors to reconstruct GRNs from time-course scRNA-seq data. dynDeepDRIM can effectively remove the transitive TF-gene interactions by considering neighborhood context and model the gene expression dynamics using high-dimensional tensors. We compared dynDeepDRIM with six GRN reconstruction methods on both simulation and four real time-course scRNA-seq data. dynDeepDRIM achieved substantially better performance than the other methods in inferring TF-gene interactions and eliminated the false positives effectively. We also applied dynDeepDRIM to annotate gene functions and found it achieved evidently better performance than the other tools due to considering the neighbor genes.

Approximately 50% of Alzheimer's disease (AD) patients will develop psychotic symptoms and these patients will experience severe rapid cognitive decline compared with those without psychosis (AD-P). Currently, no medication has been approved by the Food and Drug Administration for AD with psychosis (AD+P) specifically, although atypical antipsychotics are widely used in clinical practice. These drugs have demonstrated modest efficacy in managing psychosis in individuals with AD, with an increased frequency of adverse events, including excess mortality. We compared the differences between the genetic variations/genes associated with AD+P and schizophrenia from existing Genome-Wide Association Study and differentially expressed genes (DEGs). We also constructed disease-specific protein-protein interaction networks for AD+P and schizophrenia. Network efficiency was then calculated to characterize the topological structures of these two networks. The efficiency of antipsychotics in these two networks was calculated. A weight adjustment based on binding affinity to drug targets was later applied to refine our results, and 2013 and 2123 genes were identified as related to AD+P and schizophrenia, respectively, with only 115 genes shared. Antipsychotics showed a significantly lower efficiency in the AD+P network than in the schizophrenia network (P < 0.001) indicating that antipsychotics may have less impact in AD+P than in schizophrenia. AD+P may be caused by mechanisms distinct from those in schizophrenia which result in a decreased efficacy of antipsychotics in AD+P. In addition, the network analysis methods provided quantitative explanations of the lower efficacy of antipsychotics in AD+P.

With the development of research on the complex aetiology of many diseases, computational drug repositioning methodology has proven to be a shortcut to costly and inefficient traditional methods. Therefore, developing more promising computational methods is indispensable for finding new candidate diseases to treat with existing drugs. In this paper, a model integrating a new variant of message passing neural network and a novel-gated fusion mechanism called GLGMPNN is proposed for drug-disease association prediction. First, a light-gated message passing neural network (LGMPNN), including message passing, aggregation and updating, is proposed to separately extract multiple pieces of information from the similarity networks and the association network. Then, a gated fusion mechanism consisting of a forget gate and an output gate is applied to integrate the multiple pieces of information to extent. The forget gate calculated by the multiple embeddings is built to integrate the association information into the similarity information. Furthermore, the final node representations are controlled by the output gate, which fuses the topology information of the networks and the initial similarity information. Finally, a bilinear decoder is adopted to reconstruct an adjacency matrix for drug-disease associations. Evaluated by 10-fold cross-validations, GLGMPNN achieves excellent performance compared with the current models. The following studies show that our model can effectively discover novel drug-disease associations.

The unique chemical reactivity of cysteine residues results in various posttranslational modifications (PTMs), which are implicated in regulating a range of fundamental biological processes. With the advent of chemical proteomics technology, thousands of cysteine PTM (CysPTM) sites have been identified from multiple species. A few CysPTM-based databases have been developed, but they mainly focus on data collection rather than various annotations and analytical integration. Here, we present a platform-dubbed CysModDB, integrated with the comprehensive CysPTM resources and analysis tools. CysModDB contains five parts: (1) 70 536 experimentally verified CysPTM sites with annotations of sample origin and enrichment techniques, (2) 21 654 modified proteins annotated with functional regions and structure information, (3) cross-references to external databases such as the protein-protein interactions database, (4) online computational tools for predicting CysPTM sites and (5) integrated analysis tools such as gene enrichment and investigation of sequence features. These parts are integrated using a customized graphic browser and a Basket. The browser uses graphs to represent the distribution of modified sites with different CysPTM types on protein sequences and mapping these sites to the protein structures and functional regions, which assists in exploring cross-talks between the modified sites and their potential effect on protein functions. The Basket connects proteins and CysPTM sites to the analysis tools. In summary, CysModDB is an integrated platform to facilitate the CysPTM research, freely accessible via https://cysmoddb.bioinfogo.org/.

More and more evidence indicates that the dysregulations of microRNAs (miRNAs) lead to diseases through various kinds of underlying mechanisms. Identifying the multiple types of disease-related miRNAs plays an important role in studying the molecular mechanism of miRNAs in diseases. Moreover, compared with traditional biological experiments, computational models are time-saving and cost-minimized. However, most tensor-based computational models still face three main challenges: (i) easy to fall into bad local minima; (ii) preservation of high-order relations; (iii) false-negative samples. To this end, we propose a novel tensor completion framework integrating self-paced learning, hypergraph regularization and adaptive weight tensor into nonnegative tensor factorization, called SPLDHyperAWNTF, for the discovery of potential multiple types of miRNA-disease associations. We first combine self-paced learning with nonnegative tensor factorization to effectively alleviate the model from falling into bad local minima. Then, hypergraphs for miRNAs and diseases are constructed, and hypergraph regularization is used to preserve the high-order complex relations of these hypergraphs. Finally, we innovatively introduce adaptive weight tensor, which can effectively alleviate the impact of false-negative samples on the prediction performance. The average results of 5-fold and 10-fold cross-validation on four datasets show that SPLDHyperAWNTF can achieve better prediction performance than baseline models in terms of Top-1 precision, Top-1 recall and Top-1 F1. Furthermore, we implement case studies to further evaluate the accuracy of SPLDHyperAWNTF. As a result, 98 (MDAv2.0) and 98 (MDAv2.0-2) of top-100 are confirmed by HMDDv3.2 dataset. Moreover, the results of enrichment analysis illustrate that unconfirmed potential associations have biological significance.

Multiple types of non-canonical nucleic acid structures play essential roles in DNA recombination and replication, transcription, and genomic instability and have been associated with several human diseases. Thus, an increasing number of experimental and bioinformatics methods have been developed to identify these structures. To date, most reviews have focused on the features of non-canonical DNA/RNA structure formation, experimental approaches to mapping these structures, and the association of these structures with diseases. In addition, two reviews of computational algorithms for the prediction of non-canonical nucleic acid structures have been published. One of these reviews focused only on computational approaches for G4 detection until 2020. The other mainly summarized the computational tools for predicting cruciform, H-DNA and Z-DNA, in which the algorithms discussed were published before 2012. Since then, several experimental and computational methods have been developed. However, a systematic review including the conformation, sequencing mapping methods and computational prediction strategies for these structures has not yet been published. The purpose of this review is to provide an updated overview of conformation, current sequencing technologies and computational identification methods for non-canonical nucleic acid structures, as well as their strengths and weaknesses. We expect that this review will aid in understanding how these structures are characterised and how they contribute to related biological processes and diseases.

Discovering the biological basis of aging is one of the greatest remaining challenges for biomedical field. Work on the biology of aging has discovered a range of interventions and pathways that control aging rate. Thus, we developed AgingBank (http://bio-bigdata.hrbmu.edu.cn/AgingBank) which was a manually curated comprehensive database and high-throughput analysis platform that provided experimentally supported multi-omics data relevant to aging in multiple species. AgingBank contained 3771 experimentally verified aging-related multi-omics entries from studies across more than 50 model organisms, including human, mice, worms, flies and yeast. The records included genome (single nucleotide polymorphism, copy number variation and somatic mutation), transcriptome [mRNA, long non-coding RNA (lncRNA), microRNA (miRNA) and circular RNA (circRNA)], epigenome (DNA methylation and histone modification), other modification and regulation elements (transcription factor, enhancer, promoter, gene silence, alternative splicing and RNA editing). In addition, AgingBank was also an online computational analysis platform containing five useful tools (Aging Landscape, Differential Expression Analyzer, Data Heat Mapper, Co-Expression Network and Functional Annotation Analyzer), nearly 112 high-throughput experiments of genes, miRNAs, lncRNAs, circRNAs and methylation sites related with aging. Cancer & Aging module was developed to explore the relationships between aging and cancer. Submit & Analysis module allows users upload and analyze their experiments data. AginBank is a valuable resource for elucidating aging-related biomarkers and relationships with other diseases.

Accurately identifying cell-populations is paramount to the quality of downstream analyses and overall interpretations of single-cell RNA-seq (scRNA-seq) datasets but remains a challenge. The quality of single-cell clustering depends on the proximity metric used to generate cell-to-cell distances. Accordingly, proximity metrics have been benchmarked for scRNA-seq clustering, typically with results averaged across datasets to identify a highest performing metric. However, the 'best-performing' metric varies between studies, with the performance differing significantly between datasets. This suggests that the unique structural properties of an scRNA-seq dataset, specific to the biological system under study, have a substantial impact on proximity metric performance. Previous benchmarking studies have omitted to factor the structural properties into their evaluations. To address this gap, we developed a framework for the in-depth evaluation of the performance of 17 proximity metrics with respect to core structural properties of scRNA-seq data, including sparsity, dimensionality, cell-population distribution and rarity. We find that clustering performance can be improved substantially by the selection of an appropriate proximity metric and neighbourhood size for the structural properties of a dataset, in addition to performing suitable pre-processing and dimensionality reduction. Furthermore, popular metrics such as Euclidean and Manhattan distance performed poorly in comparison to several lessor applied metrics, suggesting that the default metric for many scRNA-seq methods should be re-evaluated. Our findings highlight the critical nature of tailoring scRNA-seq analyses pipelines to the dataset under study and provide practical guidance for researchers looking to optimize cell-similarity search for the structural properties of their own data.

Breast cancer patients often have recurrence and metastasis after surgery. Predicting the risk of recurrence and metastasis for a breast cancer patient is essential for the development of precision treatment. In this study, we proposed a novel multi-modal deep learning prediction model by integrating hematoxylin & eosin (H&E)-stained histopathological images, clinical information and gene expression data. Specifically, we segmented tumor regions in H&E into image blocks (256 × 256 pixels) and encoded each image block into a 1D feature vector using a deep neural network. Then, the attention module scored each area of the H&E-stained images and combined image features with clinical and gene expression data to predict the risk of recurrence and metastasis for each patient. To test the model, we downloaded all 196 breast cancer samples from the Cancer Genome Atlas with clinical, gene expression and H&E information simultaneously available. The samples were then divided into the training and testing sets with a ratio of 7: 3, in which the distributions of the samples were kept between the two datasets by hierarchical sampling. The multi-modal model achieved an area-under-the-curve value of 0.75 on the testing set better than those based solely on H&E image, sequencing data and clinical data, respectively. This study might have clinical significance in identifying high-risk breast cancer patients, who may benefit from postoperative adjuvant treatment.

BACKGROUND: Network medicine is an emerging area of research that focuses on delving into the molecular complexity of the disease, leading to the discovery of network biomarkers and therapeutic target discovery. Amyotrophic lateral sclerosis (ALS) is a complicated rare disease with unknown pathogenesis and no available treatment. In ALS, network properties appear to be potential biomarkers that can be beneficial in disease-related applications when explored independently or in tandem with machine learning (ML) techniques. OBJECTIVE: This systematic literature review explores recent trends in network medicine and implementations of network-based ML algorithms in ALS. We aim to provide an overview of the identified primary studies and gather details on identifying the potential biomarkers and delineated pathways. METHODS: The current study consists of searching for and investigating primary studies from PubMed and Dimensions.ai, published between 2018 and 2022 that reported network medicine perspectives and the coupling of ML techniques. Each abstract and full-text study was individually evaluated, and the relevant studies were finally included in the review for discussion once they met the inclusion and exclusion criteria. RESULTS: We identified 109 eligible publications from primary studies representing this systematic review. The data coalesced into two themes: application of network science to identify disease modules and promising biomarkers in ALS, along with network-based ML approaches. Conclusion This systematic review gives an overview of the network medicine approaches and implementations of network-based ML algorithms in ALS to determine new disease genes, and identify critical pathways and therapeutic target discovery for personalized treatment.

We performed systematic assessment of computational deconvolution methods that play an important role in the estimation of cell type proportions from bulk methylation data. The proposed framework methylDeConv (available as an R package) integrates several deconvolution methods for methylation profiles (Illumina HumanMethylation450 and MethylationEPIC arrays) and offers different cell-type-specific CpG selection to construct the extended reference library which incorporates the main immune cell subsets, epithelial cells and cell-free DNAs. We compared the performance of different deconvolution algorithms via simulations and benchmark datasets and further investigated the associations of the estimated cell type proportions to cancer therapy in breast cancer and subtypes in melanoma methylation case studies. Our results indicated that the deconvolution based on the extended reference library is critical to obtain accurate estimates of cell proportions in non-blood tissues.