embryoTox: Using Graph-Based Signatures to Predict the Teratogenicity of Small Molecules

piscesCSM: prediction of anticancer synergistic drug combinations

While drug combination therapies are of great importance, particularly in cancer treatment, identifying novel synergistic drug combinations has been a challenging venture. Computational methods have emerged in this context as a promising tool for prioritizing drug combinations for further evaluation, though they have presented limited performance, utility, and interpretability. Here, we propose a novel predictive tool, piscesCSM, that leverages graph-based representations to model small molecule chemical structures to accurately predict drug combinations with favourable anticancer synergistic effects against one or multiple cancer cell lines. Leveraging these insights, we developed a general supervised machine learning model to guide the prediction of anticancer synergistic drug combinations in over 30 cell lines. It achieved an area under the receiver operating characteristic curve (AUROC) of up to 0.89 on independent non-redundant blind tests, outperforming state-of-the-art approaches on both large-scale oncology screening data and an independent test set generated by AstraZeneca (with more than a 16% improvement in predictive accuracy). Moreover, by exploring the interpretability of our approach, we found that simple physicochemical properties and graph-based signatures are predictive of chemotherapy synergism. To provide a simple and integrated platform to rapidly screen potential candidate pairs with favourable synergistic anticancer effects, we made piscesCSM freely available online at https://biosig.lab.uq.edu.au/piscescsm/ as a web server and API. We believe that our predictive tool will provide a valuable resource for optimizing and augmenting combinatorial screening libraries to identify effective and safe synergistic anticancer drug combinations. SCIENTIFIC CONTRIBUTION: This work proposes piscesCSM, a machine-learning-based framework that relies on well-established graph-based representations of small molecules to identify and provide better predictive accuracy of syngenetic drug combinations. Our model, piscesCSM, shows that combining physiochemical properties with graph-based signatures can outperform current architectures on classification prediction tasks. Furthermore, implementing our tool as a web server offers a user-friendly platform for researchers to screen for potential synergistic drug combinations with favorable anticancer effects against one or multiple cancer cell lines.

Reproductive chemical database: a curated database of chemicals that modulate protein targets regulating important reproductive biological processes

Recent studies have shifted the spotlight from adult disease to gametogenesis and embryo developmental events, and these are greatly affected by various environmental chemicals, such as drugs, metabolites, pollutants, and others. Growing research has highlighted the critical importance of identifying and understanding the roles of chemicals in reproductive biology. However, the functions and mechanisms of chemicals in reproductive processes remain incomplete. We developed a comprehensive database called the Reproductive Chemical Database (RCDB) ( https://yu.life.sjtu.edu.cn/ChenLab/RCDB ) to facilitate research on chemicals in reproductive biology. This resource is founded on rigorous manual literature extraction and precise protein target prediction methodologies. This database focuses on the delineation of chemicals associated with phenotypes, diseases, or endpoints intricately associated with four important reproductive processes: female and male gamete generation, fertilization, and embryo development in human and mouse. The RCDB encompasses 93 sub-GO processes, and it revealed 1447 intricate chemical-biological process interactions. To date, the RCDB has meticulously cataloged and annotated 830 distinct chemicals, while also predicting 614 target proteins from a selection of 3800 potential candidates. Additionally, the RCDB offers an online predictive tool that empowers researchers to ascertain whether specific chemicals play discernible functional roles in these reproductive processes. The RCDB is an exhaustive, cross-platform, manually curated database, which provides a user-friendly interface to search, browse, and use reproductive processes modulators and their comprehensive related information. The RCDB will help researchers to understand the whole reproductive process and related diseases and it has the potential to promote reproduction research in the pharmacological and pathophysiological areas.

FetoML: Interpretable predictions of the fetotoxicity of drugs based on machine learning approaches

Pregnant females may use medications to manage health problems that develop during pregnancy or that they had prior to pregnancy. However, using medications during pregnancy has a potential risk to the fetus. Assessing the fetotoxicity of drugs is essential to ensure safe treatments, but the current process is challenged by ethical issues, time, and cost. Therefore, the need for in silico models to efficiently assess the fetotoxicity of drugs has recently emerged. Previous studies have proposed successful machine learning models for fetotoxicity prediction and even suggest molecular substructures that are possibly associated with fetotoxicity risks or protective effects. However, the interpretation of the decisions of the models on fetotoxicity prediction for each drug is still insufficient. This study constructed machine learning-based models that can predict the fetotoxicity of drugs while providing explanations for the decisions. For this, permutation feature importance was used to identify the general features that the model made significant in predicting the fetotoxicity of drugs. In addition, features associated with fetotoxicity for each drug were analyzed using the attention mechanism. The predictive performance of all the constructed models was significantly high (AUROC: 0.854-0.974, AUPR: 0.890-0.975). Furthermore, we conducted literature reviews on the predicted important features and found that they were highly associated with fetotoxicity. We expect that our model will benefit fetotoxicity research by providing an evaluation of fetotoxicity risks for drugs or drug candidates, along with an interpretation of that prediction.

Tracing the pathways and mechanisms involved in medicinal uses of flaxseed with computational methods and bioinformatics tools

Pharmacological drugs targeting specific pathways involved in various diseases have seen recent advancement with newer and more efficient emerging drug targets, but these drugs are limited in terms of their side effects and patient adherence. The potential of plant-based diets in the form of functional foods is increasingly being realized as an option to treat and/or prevent several diseases. In this work, we have selected flaxseed (Linum usitatissimum), also known as linseed, to study its pharmacological efficacy and proposed mechanisms of action for medicinal purposes. The target genes of linseed with Disease Specificity Index (DSI >0.6) are compared to the associated genes of diabetes mellitus, decrease in appetite, addictive behavior, cardiovascular diseases (CVDs), inflammatory bowel diseases (IBDs), and Polycystic Ovary Syndrome (PCOS), and the selected genes are further evaluated using in silico methods. The binding affinity of flaxseed to three common target proteins (CCDC28b, PDCD6IP, and USP34) is assessed by docking and molecular dynamics (MD) simulations. The results show that linseed is safe to use for mutagenic toxicity and other cardiotoxicity measures, but linseed is unsafe for embryotoxicity, hERG toxicity, and cardiac failure. The analysis of the protein-protein interaction (PPI) network, Gene Ontology (GO), and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways indicates that flaxseed can be used as a medicinal herb for treatment of diabetes mellitus, cardiovascular diseases, IBDs, and PCOS.

AI-Driven Enhancements in Drug Screening and Optimization

The greatest challenge in drug discovery remains the high rate of attrition across the different phases of the process, which cost the industry billions of dollars every year. While all phases remain crucial to ensure pharmaceutical-level safety, quality, and efficacy of the end product, streamlining these efforts toward compounds with success potential is pivotal for a more efficient and cost-effective process. The use of artificial intelligence (AI) within the pharmaceutical industry aims at just this, and has applications in preclinical screening for biological activity, optimization of pharmacokinetic properties for improved drug formulation, early toxicity prediction which reduces attrition, and pre-emptively screening for genetic changes in the biological target to improve therapeutic longevity. Here, we present a series of in silico tools that address these applications in small molecule development and describe how they can be embedded within the current pharmaceutical development pipeline.

FP-MAP: an extensive library of fingerprint-based molecular activity prediction tools

Discovering new drugs for disease treatment is challenging, requiring a multidisciplinary effort as well as time, and resources. With a view to improving hit discovery and lead compound identification, machine learning (ML) approaches are being increasingly used in the decision-making process. Although a number of ML-based studies have been published, most studies only report fragments of the wider range of bioactivities wherein each model typically focuses on a particular disease. This study introduces FP-MAP, an extensive atlas of fingerprint-based prediction models that covers a diverse range of activities including neglected tropical diseases (caused by viral, bacterial and parasitic pathogens) as well as other targets implicated in diseases such as Alzheimer's. To arrive at the best predictive models, performance of ≈4,000 classification/regression models were evaluated on different bioactivity data sets using 12 different molecular fingerprints. The best performing models that achieved test set AUC values of 0.62-0.99 have been integrated into an easy-to-use graphical user interface that can be downloaded from https://gitlab.com/vishsoft/fpmap.