Artificial Intelligence in Molecular Drug Discovery: Advances in Computational Pharmacology, Rational Drug Design, and Translational Therapeutics

Abstract

Artificial intelligence is increasingly transforming molecular drug discovery by enabling data-driven target identification, molecular representation, structure prediction, virtual screening, de novo design, lead optimization, pharmacokinetic modelling, toxicity prediction, drug repurposing, and preclinical translation. This review examines the methodological foundations and therapeutic applications of machine learning, deep learning, graph neural networks, generative models, network pharmacology, quantitative structure–activity relationship modelling, and multimodal computational platforms across the drug-development continuum. AI systems can integrate chemical, structural, multi-omics, pharmacological, and clinical data to prioritize disease-relevant targets, explore ultra-large chemical spaces, predict protein–ligand interactions, optimize candidate properties, and identify efficacy or safety liabilities earlier. Their application may reduce experimental burden, improve decision-making, and support precision therapeutics by linking molecular characteristics with patient-specific response patterns. Translation remains constrained by incomplete and biased datasets, inconsistent assay standards, weak external validation, limited interpretability, uncertain causal relevance, and poor generalizability across populations and disease settings. Computational predictions cannot replace experimental pharmacology, medicinal chemistry, human-relevant models, or clinical evaluation. Future progress requires explainable and uncertainty-aware models, standardized benchmarks, prospective validation, federated data infrastructures, multimodal foundation models, and closed-loop integration with automated synthesis and biological testing. Rigorous governance and multidisciplinary collaboration will determine whether AI delivers safer, more efficient, and clinically meaningful therapeutic development for diverse diseases and unmet medical needs worldwide.