Answers are grounded in this paper and cite their sources.
Review reveals AI's role in optimizing drug design and improving toxicology predictions in drug development.
Traditional technologies for new drug research and development face numerous challenges, such as the lack of high-success-rate tools for virtual design of drug molecules based on protein three-dimensional structures (a limitation to research and development efficiency), the absence of reliable methods for generating new scaffold drug molecules, and a shortage of fast, reliable, low-cost models for drug toxicology prediction. AI can accurately predict protein structures to accelerate target design, efficiently generate new scaffold molecules adapted to targets, and greatly enhance the generalization ability of protein-ligand binding predictions. Some AI models have shown excellent performance in pharmacotoxicology prediction and treatment evaluation. These new technologies significantly shorten research and development cycles, reduce costs, improve prediction accuracy and efficiency, and drive the transformation of new drug research and development from experience-driven to data-driven approaches. This article reviews the application status and progress of AI tools in drug research and development, which focuses on two areas: AI-driven cancer drug target identification and optimization, and toxicology prediction and evaluation tools.
No takes yet. Share an insight, caveat, or question.
Xinran Wang (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: