Chapter 10 - Machine learning-enforced bioinformatics approaches for drug discovery and development

Abstract

Drug resistance in microbes and parasites has extended extensively all over the globe, necessitating the discovery of novel therapeutics and drug molecules for the cures of many infectious and debilitating diseases that affect humans. Therapeutic drug interactions with microbial or pathogenic cell receptors are essential for lowering or preventing cellular growth. Numerous structural and morphological investigation bioinformatic tools have been used to comprehend the mechanism of action of ligand and protein interactions. Integration of machine learning (ML) with bioinformatics tools is the most attention attraction in drug discovery, repurposing, and delivery technology. ML potentially improved biological data analysis, decision-making, drug discovery, and preclinical studies by reducing cost and time. Enforcement of ML with bioinformatics has established standard, robust, and reproducible computational approaches to achieve the goals. Especially deep ML has become a game-changer in the medical sector, predominantly drug discovery. This chapter elaborates an in-depth study of numerous ML strategies or models integrated with bioinformatics, their development, frameworks, workflows, state-of-the-art contribution in cheminformatics of drug discovery applications, and briefing the limitations of ML, including their future perceptions.