Artificial Intelligence and Machine Learning in Drug Design and Development

4Machine Learning in Drug Discovery: Methods, Applications, and Challenges

Geetha Mani¹* and Gokulakrishnan Jayakumar²

¹Department of Control and Automation, School of Electrical Engineering, Vellore Institute of Technology, Vellore Campus, Vellore, Tamil Nadu, India

²School for Environment and Sustainability, University of Michigan – Ann Arbor, Ann Arbor, MI, United States

Abstract

In recent times, the pharmaceutical industry has leveraged cutting-edge technologies to gather insights from intricate biological systems. Advances in technology have spurred a growing interest in applying machine learning (ML) and deep learning (DL) approaches to generate potential therapeutic hypotheses for drug discovery and development (DD&D). These approaches offer a suite of frameworks and accompanying toolkits that enhance the discovery and decision-making processes for well-defined problems using diverse, large-scale data. In this domain, ML algorithms can be integrated across all phases of DD&D. Among their many potential applications are target validation, the identification of prognostic biomarkers, the predictive analytics of drug–protein interactions and drug toxicity, and the data analysis of digital pathology in clinical trials. This chapter delivers a comprehensive overview of both unsupervised and supervised ML algorithms that drug discovery scientists and researchers have employed, along with their advantageous applications. Utilizing ML algorithms in drug discovery and development ...