Editors: Dr Mukul Kumar Baruah, Dr Rahul Kanti Nath & Dr Joyobrato Nath (2025)

ISBN: 97978-93-94174-62-7

Machine Learning Approaches in Drug-design: Application of Artificial Intelligence to Predict Chemical Structure-Biological Activity Relationship

Department of Chemistry, University of Science and Technology Meghalaya, Meghalaya 793101, India

Recent advancements in artificial intelligence (AI) and machine learning (ML) have enabled the establishment of robust correlations between chemical structure and biological activity, known as quantitative structure–activity relationships (QSARs). These models utilize chemical descriptors to predict the biological or hazardous activity of compounds, especially in scenarios where experimental data is scarce, thus aiding in the prioritization of chemicals for further testing and regulatory evaluation. This review compiles current and emerging literature on the application of ML tools and techniques in drug discovery and development. ML methods are increasingly being integrated across all phases of the drug development pipeline, aiming to accelerate research, reduce clinical trial costs, and minimize risk. ML enhances decision-making by processing complex pharmaceutical datasets, enabling advances in areas such as QSAR modeling, hit identification, and de novo drug design. These data-driven approaches offer more precise and predictive outcomes, supporting early-stage screening and lead optimization. The review emphasizes the need for high-quality, structured data to support the robust validation of ML models. Addressing issues related to methodological rigor, model reliability, and interpretability is essential to enhance confidence in ML-driven decisions. Furthermore, raising awareness and understanding of ML among stakeholders will be key to reduce failure rates, improving trial success, and expediting drug development processes..

*Corresponding author; e-mail: samiyara.ustm@gmail.com

Cite this Chapter: Samiyara Begum. 2025. Machine Learning Approaches in Drug-design: Application of Artificial Intelligence to Predict Chemical Structure-Biological Activity Relationship. In: Mukul Kumar Baruah, Rahul Kanti Nath and Joyobrato Nath (Eds.), Transdisciplinary Science: Mapping the Future of Research. Excellent Publishers, India. pp. 206-218. doi: https://doi.org/10.20546/978-93-94174-62-7_19