Artificial Intelligence in Medicinal Chemistry: A Comprehensive Review

Abstract

Artificial intelligence (AI) has developed as a transformative force in medicinal chemistry, redesigning early drug discovery, hit identification, lead optimisation, and synthetic route planning. Effective exploration of chemical space and determination of key molecular properties has been enabled by advances in machine learning (ML), deep learning (DL), graph neural networks (GNNs), and generative models (such as variational autoencoders and transformer-based chemical language models). AI now plays a vital role across virtual screening, de-novo design, reaction prediction, retrosynthesis planning, ADMET profiling, and structure-based drug design. Remains challenging despite its rapid progress which includes data quality issues, model interpretability, experimental validation gaps, and integration into existing medicinal chemistry workflows. This review summarises the current AI methodologies, major applications, advantages, limitations and future opportunities in medicinal chemistry.