Motivation: Drug combination therapy has become an increasingly promising method in the treatment of cancer. However, the number of possible drug combinations is so huge that it is hard to screen synergistic drug combinations through wet-lab experiments. Therefore, computational screening has become an important way to prioritize drug combinations. Graph neural network has recently shown remarkable performance in the prediction of compound–protein interactions, but it has not been applied to the screening of drug combinations. Results: In this paper, we proposed a deep learning model based on graph neural network and attention mechanism to identify drug combinations that can effectively inhibit the viability of specific cancer cells. The feature embeddings of drug molecule structure and gene expression profiles were taken as input to multilayer feedforward neural network to identify the synergistic drug combinations. We compared DeepDDS (Deep Learning for Drug–Drug Synergy prediction) with classical machine learning methods and other deep learning-based methods on benchmark data set, and the leave-one-out experimental results showed that DeepDDS achieved better performance than competitive methods. Also, on an independent test set released by well-known pharmaceutical enterprise AstraZeneca, DeepDDS was superior to competitive methods by more than 16% predictive precision. Furthermore, we explored the interpretability of the graph attention network and found the correlation matrix of atomic features revealed important chemical substructures of drugs. We believed that DeepDDS is an effective tool that prioritized synergistic drug combinations for further wet-lab experiment validation. Availability and implementation: Source code and data are available at https://github.com/Sinwang404/DeepDDS/tree/master

Synthetic lethality (SL) is a type of genetic interaction between two genes such that simultaneous perturbations of the two genes result in cell death or a dramatic decrease of cell viability, while a perturbation of either gene alone is not lethal. SL reflects the biologically endogenous difference between cancer cells and normal cells, and thus the inhibition of SL partners of genes with cancer-specific mutations could selectively kill cancer cells but spare normal cells. Therefore, SL is emerging as a promising anticancer strategy that could potentially overcome the drawbacks of traditional chemotherapies by reducing severe side effects. Researchers have developed experimental technologies and computational prediction methods to identify SL gene pairs on human and a few model species. However, there has not been a comprehensive database dedicated to collecting SL pairs and related knowledge. In this paper, we propose a comprehensive database, SynLethDB (http://histone.sce.ntu.edu.sg/SynLethDB/), which contains SL pairs collected from biochemical assays, other related databases, computational predictions and text mining results on human and four model species, i.e. mouse, fruit fly, worm and yeast. For each SL pair, a confidence score was calculated by integrating individual scores derived from different evidence sources. We also developed a statistical analysis module to estimate the druggability and sensitivity of cancer cells upon drug treatments targeting human SL partners, based on large-scale genomic data, gene expression profiles and drug sensitivity profiles on more than 1000 cancer cell lines. To help users access and mine the wealth of the data, we developed other practical functionalities, such as search and filtering, orthology search, gene set enrichment analysis. Furthermore, a user-friendly web interface has been implemented to facilitate data analysis and interpretation. With the integrated data sets and analytics functionalities, SynLethDB would be a useful resource for biomedical research community and pharmaceutical industry.

Since traditional drug research and development is often time-consuming and high-risk, there is an increasing interest in establishing new medical indications for approved drugs, referred to as drug repositioning, which provides a relatively low-cost and high-efficiency approach for drug discovery. With the explosive growth of large-scale biochemical and phenotypic data, drug repositioning holds great potential for precision medicine in the post-genomic era. It is urgent to develop rational and systematic approaches to predict new indications for approved drugs on a large scale. In this paper, we propose the two-pass random walks with restart on a heterogenous network, TP-NRWRH for short, to predict new indications for approved drugs. Rather than random walk on bipartite network, we integrated the drug-drug similarity network, disease-disease similarity network and known drug-disease association network into one heterogenous network, on which the two-pass random walks with restart is implemented. We have conducted performance evaluation on two datasets of drug-disease associations, and the results show that our method has higher performance than six existing methods. A case study on the Alzheimer’s disease showed that nine of top 10 predicted drugs have been approved or investigational for neurodegenerative diseases. The experimental results show that our method achieves state-of-the-art performance in predicting new indications for approved drugs. We proposed a two-pass random walk with restart on the drug-disease heterogeneous network, referred to as TP-NRWRH, to predict new indications for approved drugs. Performance evaluation on two independent datasets showed that TP-NRWRH achieved higher performance than six existing methods on 10-fold cross validations. The case study on the Alzheimer’s disease showed that nine of top 10 predicted drugs have been approved or are investigational for neurodegenerative diseases. The results show that our method achieves state-of-the-art performance in predicting new indications for approved drugs.

Two genes are synthetic lethal if mutations in both genes result in impaired cell viability, while mutation of either gene does not affect the cell survival. The potential usage of synthetic lethality (SL) in anticancer therapeutics has attracted many researchers to identify synthetic lethal gene pairs. To include newly identified SLs and more related knowledge, we present a new version of the SynLethDB database to facilitate the discovery of clinically relevant SLs. We extended the first version of SynLethDB database significantly by including new SLs identified through Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR) screening, a knowledge graph about human SLs, a new web interface, etc. Over 16 000 new SLs and 26 types of other relationships have been added, encompassing relationships among 14 100 genes, 53 cancers, 1898 drugs, etc. Moreover, a brand-new web interface has been developed to include modules such as SL query by disease or compound, SL partner gene set enrichment analysis and knowledge graph browsing through a dynamic graph viewer. The data can be downloaded directly from the website or through the RESTful Application Programming Interfaces (APIs). Database URL: https://synlethdb.sist.shanghaitech.edu.cn/v2.

Accumulated clinical studies show that microbes living in humans interact closely with human hosts, and get involved in modulating drug efficacy and drug toxicity. Microbes have become novel targets for the development of antibacterial agents. Therefore, screening of microbe–drug associations can benefit greatly drug research and development. With the increase of microbial genomic and pharmacological datasets, we are greatly motivated to develop an effective computational method to identify new microbe–drug associations. In this article, we proposed a novel method, Graph2MDA, to predict microbe–drug associations by using variational graph autoencoder (VGAE). We constructed multi-modal attributed graphs based on multiple features of microbes and drugs, such as molecular structures, microbe genetic sequences and function annotations. Taking as input the multi-modal attribute graphs, VGAE was trained to learn the informative and interpretable latent representations of each node and the whole graph, and then a deep neural network classifier was used to predict microbe–drug associations. The hyperparameter analysis and model ablation studies showed the sensitivity and robustness of our model. We evaluated our method on three independent datasets and the experimental results showed that our proposed method outperformed six existing state-of-the-art methods. We also explored the meaning of the learned latent representations of drugs and found that the drugs show obvious clustering patterns that are significantly consistent with drug ATC classification. Moreover, we conducted case studies on two microbes and two drugs and found 75–95% predicted associations have been reported in PubMed literature. Our extensive performance evaluations validated the effectiveness of our proposed method. Source codes and preprocessed data are available at https://github.com/moen-hyb/Graph2MDA. Supplementary data are available at Bioinformatics online.