Advancing Human and Environmental Safety Science Using In Silico Methods

Advancing Human and Environmental Safety Science Using In Silico Methods Cite This: Chem. Res. Toxicol. 2025, 38, 1281−1282 Read Online ACCESS Metrics & More Article Recommendations The 21st edition of the International Workshop on Quantitative Structure−Activity Relationships in Environmental and Health Sciences (QSAR2025) took place in Milan, Italy, on June 3−6, 2025 (Figure 1). This pivotal event fostered innovation in chemical safety by uniting scientists and stakeholders from different sectors (academia, research institutes, industry and regulatory agencies) to discuss the development and application of in silico methods in human and environmental health. Computational (in silico) methods, like (Q)SAR and read-across, are non-animal testing approaches that fall under the umbrella of New Approach Methodologies (NAMs)�also including in vitro,in chemico, omics, etc., which offer the opportunity to leverage and integrate diverse knowledge to inform chemical hazard and risk assessment. The QSAR2025 conference provided insights and perspectives on emerging research findings, tools, and regulatory applications related to in silico methods. Several diverse, innovative research areas were covered during the presentations, including federated learning, AI driven approaches, case studies on the use of in silico methods for regulatory purposes, new QSAR predictive tools and approaches, to name a few. In addition to talks from academic partners, valuable contributions from representatives of regulatory agencies, large industrial companies, Organisation for Economic Co-operation and Development (OECD) and many small companies and consultancies actively working in the field were featured in the program. The topics covered during the workshop included the following areas of research: •AI, big data, and QSAR •Methodological aspects of (Q)SAR modeling •Quantification of data and model uncertainties •In silico strategies for data-poor end points or substances •(Q)SARs for screening, prioritization and data gap filling •Implementing and integrating in silico NAMs for regulatory decision-making contexts •Read-across, grouping, and consensus modeling within a Weight of Evidence framework •(Q)SARs for toxicokinetic modeling •Emerging technologies and challenges (e.g., UVCBs, mixtures, nano) •eXplainable AI (XAI) methods •In silico approaches for ecotoxicity, green chemistry, sustainability, and climate change •Drug design and drug toxicity The current Special Issue aims to provide a comprehensive collection of articles covering all these topics presented by Workshop participants, but it is open as well to contributors from all over the world working toward in silico methodologies advancing human and environmental safety. This issue thus complements the more theoretical collections devoted to computational toxicology 1 and the use of artificial intelligence in toxicology, 2 by focusing more on practical uses of the novel methods for human and environmental safety. We are looking forward to receiving novel methodological works that represent significant advancements in the field’s state of the art by providing more accurate predictions (such as methods developed within Tox24 Challenge 3 ), which are also correctly validated and interpreted to allow for their unbiased use in prospective studies 4 and in particular for regulatory use. 5 In this respect, linking predictions to molecular initiating events (MIE) and adverse outcome pathways (AOPs) to provide mechanistic explanations is highly encouraged. 6 Comprehensive reviews covering recent developments in the field and/or perspectives discussing new advances and unexplored ideas/methods that are expected to have high impact in the (eco-)toxicology context (e.g., use of Large Language Models, LLM, to assist regulatory decisions, impact of foundation models, and generative AI methods), as well as novel developments in AOP studies are very welcome. The authors submitting work to this Special Issue should conclude the abstract with a scientific contribution statement, 2−3 sentences long (maximum), which clearly indicates how the submission advances in silico methodologies beyond the current state of the art and differentiates the submitted work from the previous studies. The special issue is open for submissions up to December 31st, 2025. Alessandra Roncaglioni Simona Kovarich Kamel Mansouri Igor V. Tetko orcid.org/0000-0002-6855-0012 Published: August 8, 2025 Editorialpubs.acs.org/crt Published 2025 by American Chemical Society 1281 https://doi.org/10.1021/acs.chemrestox.5c00293 Chem. Res. Toxicol. 2025, 38, 1281−1282 Downloaded via 185.17.204.34 on November 3, 2025 at 15:32:09 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles. ■AUTHOR INFORMATION Complete contact information is available at: https://pubs.acs.org/10.1021/acs.chemrestox.5c00293 Notes Views expressed in this editorial are those of the authors and not necessarily the views of the ACS. ■ACKNOWLEDGMENTS This editorial was partially funded by the European Union’s Horizon Europe programme under the Marie SkłodowskaCurie Actions Doctoral Networks grant agreement No. 101120466 “Explainable AI for Molecules” (AiChemist). ■REFERENCES (1) Kleinstreuer, N. C.; Tetko, I. V.; Tong, W. Introduction to Special Issue: Computational Toxicology. Chem. Res. Toxicol. 2021, 34 (2), 171−175. (2) Klambauer, G.; Clevert, D.-A.; Shah, I.; Benfenati, E.; Tetko, I. V. Introduction to the Special Issue: AI Meets Toxicology. Chem. Res. Toxicol. 2023,36 (8), 1163−1167. (3) Eytcheson, S. A.; Tetko, I. V. Which Modern AI Methods Provide Accurate Predictions of Toxicological Endpoints? Analysis of Tox24 Challenge Results. Chem. Res. Tox. 2025,DOI: 10.1021/ acs.chemrestox.5c00273. (4) Seal, S.; Mahale, M.; García-Ortegón, M.; Joshi, C. K.; HosseiniGerami, L.; Beatson, A.; Greenig, M.; Shekhar, M.; Patra, A.; Weis, C.; Mehrjou, A.; Badré, A.; Paisley, B.; Lowe, R.; Singh, S.; Shah, F.; Johannesson, B.; Williams, D.; Rouquie, D.; Clevert, D.-A.; Schwab, P.; Richmond, N.; Nicolaou, C. A.; Gonzalez, R. J.; Naven, R.; Schramm, C.; Vidler, L. R.; Mansouri, K.; Walters, W. P.; Wilk, D. D.; Spjuth, O.; Carpenter, A. E.; Bender, A. Machine Learning for Toxicity Prediction Using Chemical Structures: Pillars for Success in the Real World. Chem. Res. Toxicol. 2025,38 (5), 759−807. (5) Kovarich, S.; Cappelli, C. I. Use of In Silico Methods for Regulatory Toxicological Assessment of Pharmaceutical Impurities. In In Silico Methods for Predicting Drug Toxicity; Benfenati, E., Ed.; Springer US: New York, NY, 2022; pp 537−560. DOI: 10.1007/9781-0716-1960-5_21. (6) Gadaleta, D.; Garcia de Lomana, M.; Serrano-Candelas, E.; Ortega-Vallbona, R.; Gozalbes, R.; Roncaglioni, A.; Benfenati, E. Quantitative Structure−Activity Relationships of Chemical Bioactivity toward Proteins Associated with Molecular Initiating Events of Organ-Specific Toxicity. J. Cheminformatics 2024,16 (1), 122. Figure 1. Logo of the International Workshop on Quantitative Structure−Activity Relationships in Environmental and Health Sciences (QSAR2025). Chemical Research in Toxicology pubs.acs.org/crt Editorial https://doi.org/10.1021/acs.chemrestox.5c00293 Chem. Res. Toxicol. 2025, 38, 1281−1282 1282