Integration of Molecular Docking with Artificial Intelligence and Machine Learning for Accelerated Drug Discovery

Molecular docking has been a cornerstone of computer-aided drug discovery, enabling prediction of ligand–receptor interactions and prioritization of drug candidates. However, traditional docking approaches face challenges related to accuracy, computational cost, and limited ability to account for biological complexity. In recent years, artificial intelligence (AI) and machine learning (ML) have emerged as transformative technologies capable of addressing these limitations. Integration of molecular docking with AI/ML approaches allows not only rapid screening of vast chemical libraries but also improved prediction of binding affinities, pose selection, and off-target interactions. This review summarizes recent advances (2020–2025) in the combined application of molecular docking and AI/ML for accelerated drug discovery. Case studies highlight applications in neurodegenerative disorders, antimicrobial resistance, and oncology, where integrated approaches have yielded significant improvements in candidate identification and optimization. The article also discusses current limitations, such as data scarcity, reproducibility challenges, and interpretability of AI models. Finally, future perspectives on incorporating generative AI, multimodal data integration, and cloud-based collaborative platforms are presented. By synergizing traditional molecular docking with cutting-edge AI and ML techniques, the drug discovery pipeline can be significantly accelerated, reducing cost and increasing the probability of successful therapeutic development.