Medicine in Drug Discovery · Published 2025-08-21 · DOI 10.1016/j.medidd.2025.100223
The fusion of Artificial intelligence (AI) with computational chemistry has revolutionized drug discovery by enhancing compound optimization, predictive analytics, and molecular modeling. This review explores the integration of AI techniques, including machine learning (ML), deep learning (DL), and generative models with traditional computational methods such as molecular docking, quantum mechanics, and molecular dynamics simulations. It outlines the evolution of computational chemistry and the transformative role of AI in interpreting complex molecular data, automating feature extraction, and improving decision-making across the drug development pipeline. Core AI algorithms support vector machines, random forests, graph neural networks, and transformers are examined for their applications in molecular representation, virtual screening, and ADMET property prediction. Special attention is given to de novo drug design using generative adversarial networks (GANs) and variational autoencoders (VAEs), as well as AI-driven high-throughput virtual screening that reduces computational costs while improving hit identification. The review also discusses platforms like Deep-PK and DeepTox for pharmacokinetics and toxicity prediction using graph-based descriptors and multitask learning. In structure-based design, AI-enhanced scoring functions and binding affinity models outperform classical approaches, while DL transforms molecular dynamics by approximating force fields and capturing conformational dynamics. The convergence of AI with quantum chemistry and density functional theory (DFT) is illustrated through surrogate modeling and reaction mechanism prediction. Despite these advances, challenges remain in data quality, model interpretability, and generalizability. The review concludes by highlighting future directions, including hybrid AI-quantum frameworks and multi-omics integration, underscoring AI’s potential to accelerate safer, more cost-effective drug discovery.
Abstract from DOAJ. Public domain (CC0 1.0).
Read the article at the publisher →