Abstract

Abstract Advancements in predictive toxicology have significantly enhanced the drug development process by utilizing in silico models to assess drug safety. These computational models, including quantitative structure-activity relationship (QSAR) models, molecular docking, and machine learning algorithms, provide robust tools for predicting the toxicological effects of new compounds. In silico approaches offer substantial benefits in terms of speed, cost-efficiency, and the reduction of animal testing, enabling comprehensive toxicity assessments across various endpoints such as hepatotoxicity, cardiotoxicity, and genotoxicity. Despite challenges related to data quality, model validation, and biological complexity, continuous improvements and integration with experimental data promise to further refine these models. This review highlights the current state of in silico models in predictive toxicology, their applications in drug safety assessment, and future directions for enhancing their predictive accuracy and regulatory acceptance