Toxicity is the cause of around 1/3 attrition of drug candidates and a major contributor to cost during drug development. Predictive toxicology seeks to discover and mitigate the hazards of druggable chemicals, reduce the failure rate and the cost of a new drug development, and ultimately improve their clinical safety by integrating preclinical and clinical toxicity data with novel technologies. Machine learning can be utilised to combine chemical, biological, and mechanistic data to improve toxicity assessment of chemicals and drugs. These methods together with other deep learning approaches can make AI-based modelling a feasible alternative to traditional toxicology approaches, mitigating time and cost without raising the ethical issues of animal or clinical testing. Advanced machine learning could efficiently increase the accuracy and safety of new drug candidates in humans. However, accurately modeling a multi-faceted problem such as toxicity across distinct platforms and interpreting the vast number of predictions that would be generated remains a challenge, thereby necessitating more research to enhance human benefits.

The collection topic highlights the impact of deep learning on aiding safe drug discovery and development. We consider articles on all aspects of research that consider the potential of deep learning for refining predictive models, advancing the accuracy of chemical risk assessment, mitigating toxicity and improving the clinical efficiency and safety of drugs.

Meet our Guest Editor

Duc Nguyen, PhD, University of Kentucky, USA

Professor Duc Nguyen is an expert at the intersection of mathematics, molecular bioscience, and data science, and serves as an Associate Professor in the Department of Mathematics at the University of Kentucky. His research focuses on three areas: developing mathematical models for molecular bioscience and biophysics, designing machine learning architectures to enhance learning accuracy, and constructing high-order methods for scientific computing. His work has been supported by three NSF grants, Pfizer, and Bristol-Myers Squibb. Professor Nguyen’s significant impact is demonstrated through his success in the D3R Grand Challenges, a prestigious competition in computer-aided drug design (CADD), where his models were ranked first in several categories. Recognized among the top 2% of the world’s most-cited researchers, his contributions advance scientific understanding and innovation, specializing in math and AI-driven drug discovery.

Submission Guidelines

This Collection welcomes submission of original Research Articles. Should you wish to submit a different article type, please read our submission guidelines to confirm that type is accepted by the journal. Articles for this Collection should be submitted via our submission system, SNAPP. During the submission process you will be asked whether you are submitting to a Collection, please select “Machine learning for predictive toxicology” from the dropdown menu.

Articles will undergo the journal’s standard peer-review process and are subject to all of the journal’s standard policies. Articles will be added to the Collection as they are published.

The Editors have no competing interests with the submissions which they handle through the peer review process. The peer review of any submissions for which the Editors have competing interests is handled by another Editorial Board Member who has no competing interests.

If accepted for publication, Article Processing Charges (APC) applies. Please find out more about our standard APC waiver policy.

The Collection is now open for submissions and the deadline is 14 March 2025.