Abstract: The prediction of drug-target affinity (DTA) is a crucial step for drug screening and discovery. In this study, a new graph-based prediction model named SAG-DTA (self-attention graph drug-target affinity) was implemented. Unlike previous graph-based methods, the proposed model utilized self-attention mechanisms on the drug molecular graph to obtain effective representations of drugs for DTA prediction. Features of each atom node in the molecular graph were weighted using an attention score before being aggregated as molecule representation. Various self-attention scoring methods were compared in this study. In addition, two pooing architectures, namely, global and hierarchical architectures, were presented and evaluated on benchmark datasets. Results of comparative experiments on both regression and binary classification tasks showed that SAG-DTA was superior to previous sequence-based or other graph-based methods and exhibited good generalization ability.

Abstract: Coronavirus Disease-19 (COVID-19) has lead global epidemics with high morbidity and mortality. However, there are currently no proven effective drugs targeting COVID-19. Identifying drug-virus associations can not only provide insights into the understanding of drug-virus interaction mechanism, but also guide and facilitate the screening of compound candidates for antiviral drug discovery. Since conventional experiment methods are time-consuming, laborious and expensive, computational methods to identify potential drug candidates for viruses (e.g., COVID-19) provide an alternative strategy. In this work, we propose a novel framework of Heterogeneous Graph Attention Networks for Drug-Virus Association predictions, named HGATDVA. First, we fully incorporate multiple sources of biomedical data, e.g., drug chemical information, virus genome sequences and viral protein sequences, to construct abundant features for drugs and viruses. Second, we construct two drug-virus heterogeneous graphs. For each graph, we design a self-enhanced graph attention network (SGAT) to explicitly model the dependency between a node and its local neighbors and derive the graph-specific representations for nodes. Third, we further develop a neural network architecture with tri-aggregator to aggregate the graph-specific representations to generate the final node representations. Extensive experiments were conducted on two datasets, i.e., DrugVirus and MDAD, and the results demonstrated that our model outperformed 7 state-of-the-art methods. Case study on SARS-CoV-2 validated the effectiveness of our model in identifying potential drugs for viruses.

Abstract: Spinal muscular atrophy (SMA), one of the leading inherited causes of child mortality, is a rare neuromuscular disease arising from loss-of-function mutations of the survival motor neuron 1 (SMN1) gene, which encodes the SMN protein. When lacking the SMN protein in neurons, patients suffer from muscle weakness and atrophy, and in the severe cases, respiratory failure and death. Several therapeutic approaches show promise with human testing and three medications have been approved by the U.S. Food and Drug Administration (FDA) to date. Despite the shown promise of these approved therapies, there are some crucial limitations, one of the most important being the cost. The FDA-approved drugs are high-priced and are shortlisted among the most expensive treatments in the world. The price is still far beyond affordable and may serve as a burden for patients. The blooming of the biomedical data and advancement of computational approaches have opened new possibilities for SMA therapeutic development. This article highlights the present status of computationally aided approaches, including in silico drug repurposing, network driven drug discovery as well as artificial intelligence (AI)-assisted drug discovery, and discusses the future prospects.

Abstract: Technological advances in high-throughput techniques have resulted in tremendous growth of complex biological datasets providing evidence regarding various biomolecular interactions. To cope with this data flood, computational approaches, web services, and databases have been implemented to deal with issues such as data integration, visualization, exploration, organization, scalability, and complexity. Nevertheless, as the number of such sets increases, it is becoming more and more difficult for an end user to know what the scope and focus of each repository is and how redundant the information between them is. Several repositories have a more general scope, while others focus on specialized aspects, such as specific organisms or biological systems. Unfortunately, many of these databases are self-contained or poorly documented and maintained. For a clearer view, in this article we provide a comprehensive categorization, comparison and evaluation of such repositories for different bioentity interaction types. We discuss most of the publicly available services based on their content, sources of information, data representation methods, user-friendliness, scope and interconnectivity, and we comment on their strengths and weaknesses. We aim for this review to reach a broad readership varying from biomedical beginners to experts and serve as a reference article in the field of Network Biology.

Abstract: Members of the Bacillus genus, particularly the "Bacillus subtilis group", are known to produce amphipathic lipopeptides with biosurfactant activity. This includes the surfactins, fengycins and iturins that have been associated with antibacterial, antifungal, and anti-viral properties. We have screened a large collection of Bacillus, isolated from human, animal, estuarine water and soil samples and found that the most potent lipopeptide producers are members of the species Bacillus velezensis. B. velezensis lipopeptides exhibited anti-bacterial activity which was localised on the surface of both vegetative cells and spores. Interestingly, lipopeptide micelles (6-10 nm diameter) were detectable in strains exhibiting the highest levels of activity. Micelles were stable (heat and gastric stable) and shown to entrap other antimicrobials produced by the host bacterium (exampled here was the dipeptide antibiotic chlorotetaine). Commercially acquired lipopeptides did not exhibit similar levels of inhibitory activity and we suspect that micelle formation may relate to the particular isomeric forms produced by individual bacteria. Using naturally produced micelle formulations we demonstrated that they could entrap antimicrobial compounds (e.g., clindamycin, vancomycin and resveratrol). Micellar incorporation of antibiotics increased activity. Bacillus is a prolific producer of antimicrobials, and this phenomenon could be exploited naturally to augment antimicrobial activity. From an applied perspective, the ability to readily produce Bacillus micelles and formulate with drugs enables a possible strategy for enhanced drug delivery.

Abstract: In 2005, the NIH Molecular Libraries Program (MLP) undertook the identification of tool compounds to expand biological insights, now termed small-molecule chemical probes. This inspired other organisations to initiate similar efforts from 2010 onwards. As a central focus of the Probes & Drugs portal (P&D), we have standardised, integrated and compared sets of declared probe compounds harvested from 12 different sources. This turned out to be challenging and revealed unexpected anomalies. Results in this work address key questions including; a) individual and total structure counts, b) overlaps between sources, c) comparisons with selected PubChem sources and d) investigating the probe coverage of druggable targets. In addition, we developed new high-level scoring schemes to filter collections down to probes of higher quality. This generated 548 high-quality chemical probes (HQCP) covering 447 distinct protein targets. This HQCP collection has been added to the P&D portal and will be regularly updated as established sources expand and new ones release data.

Abstract: The return of blood flow to ischemic heart after myocardial infarction causes ischemia-reperfusion injury. There is a clinical need for novel therapeutic targets to treat myocardial ischemia-reperfusion injury. Here we screened for targets for the treatment of ischemia-reperfusion injury using a combination of shRNA and drug library analyses in HL-1 mouse cardiomyocytes subjected to hypoxia and reoxygenation. The shRNA library included lentiviral constructs targeting 4625 genes and the drug library 689 chemical compounds approved by the Food and Drug Administration (FDA). Data were analyzed using protein-protein interaction and pathway analyses. EGFR inhibition was identified as a cardioprotective mechanism in both approaches. Inhibition of EGFR kinase activity with gefitinib improved cardiomyocyte viability in vitro. In addition, gefitinib preserved cardiac contractility in zebrafish embryos exposed to hypoxia-reoxygenation in vivo. These findings indicate that the EGFR inhibitor gefitinib is a potential candidate for further studies of repurposing the drug for the treatment of myocardial infarction.

Abstract: OBJECTIVE: To summarize how artificial intelligence (AI) is being applied in COVID-19 research and determine whether these AI applications integrated heterogenous data from different sources for modeling. MATERIALS AND METHODS: We searched 2 major COVID-19 literature databases, the National Institutes of Health's LitCovid and the World Health Organization's COVID-19 database on March 9, 2021. Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guideline, 2 reviewers independently reviewed all the articles in 2 rounds of screening. RESULTS: In the 794 studies included in the final qualitative analysis, we identified 7 key COVID-19 research areas in which AI was applied, including disease forecasting, medical imaging-based diagnosis and prognosis, early detection and prognosis (non-imaging), drug repurposing and early drug discovery, social media data analysis, genomic, transcriptomic, and proteomic data analysis, and other COVID-19 research topics. We also found that there was a lack of heterogenous data integration in these AI applications. DISCUSSION: Risk factors relevant to COVID-19 outcomes exist in heterogeneous data sources, including electronic health records, surveillance systems, sociodemographic datasets, and many more. However, most AI applications in COVID-19 research adopted a single-sourced approach that could omit important risk factors and thus lead to biased algorithms. Integrating heterogeneous data for modeling will help realize the full potential of AI algorithms, improve precision, and reduce bias. CONCLUSION: There is a lack of data integration in the AI applications in COVID-19 research and a need for a multilevel AI framework that supports the analysis of heterogeneous data from different sources.

Abstract: BACKGROUND: Breast cancer (BC) is a complex disease with high heterogeneity, which often leads to great differences in treatment results. Current common molecular typing method is PAM50, which shows positive results for precision medicine; however, room for improvement still remains because of the different prognoses of subtypes. Therefore, in this article, we used lncRNAs, which are more tissue-specific and developmental stage-specific than other RNAs, as typing markers and combined single-cell expression profiles to retype BC, to provide a new method for BC classification and explore new precise therapeutic strategies based on this method. METHODS: Based on lncRNA expression profiles of 317 single cells from 11 BC patients, SC3 was used to retype BC, and differential expression analysis and enrichment analysis were performed to identify biological characteristics of new subtypes. The results were validated for survival analysis using data from TCGA. Then, the downstream regulatory genes of lncRNA markers of each subtype were searched by expression correlation analysis, and these genes were used as targets to screen therapeutic drugs, thus proposing new precision treatment strategies according to the different subtype compositions of patients. RESULTS: Seven lncRNA subtypes and their specific biological characteristics are obtained. Then, 57 targets and 210 drugs of 7 subtypes were acquired. New precision medicine strategies were proposed according to the different compositions of patient subtypes. CONCLUSIONS: For patients with different subtype compositions, we propose a strategy to select different drugs for different patients, which means using drugs targeting multi subtype or combinations of drugs targeting a single subtype to simultaneously kill different cancer cells by personalized treatment, thus reducing the possibility of drug resistance and even recurrence.

Abstract: Individual response to drugs is highly variable and largely influenced by genetic variants and gene-expression profiles. In addition, it has been shown that response to drugs is strongly sex-dependent, both in terms of efficacy and toxicity. To expand current knowledge on sex differences in the expression of genes relevant for drug response, we generated a catalogue of differentially expressed human transcripts encoded by 289 genes in 41 human tissues from 838 adult individuals of the Genotype-Tissue Expression project (GTEx, v8 release) and focused our analysis on relevant transcripts implicated in drug response. We detected significant sex-differentiated expression of 99 transcripts encoded by 59 genes in the tissues most relevant for human pharmacology (liver, lung, kidney, small intestine terminal ileum, skin not sun-exposed, and whole blood). Among them, as expected, we confirmed significant differences in the expression of transcripts encoded by the cytochromes in the liver, CYP2B6, CYP3A7, CYP3A5, and CYP1A1. Our systematic investigation on differences between male and female in the expression of drug response-related genes, reinforce the need to overcome the sex bias of clinical trials.