Evolutionary algorithms and artificial intelligence in drug discovery: opportunities, tools, and prospects

The drug design process is lengthy, complex, and dependent on several factors. Developing a medicine can take 10 to 15 years, from discovery to commercialisation. Machine learning (ML) refers to a set of tools that can assist you in learning more and making better decisions for well-defined questions with a large amount of data. The opportunities to use ML occur throughout the drug development process. Examples include target identification and validation, identification of alternative targets, and biomarker identification. Some approaches have produced accurate predictions and insights, while others have not. But to deal with high-dimensional data, we need soft-computing methods to find the best solution, which could be a new drug. This article provides a detailed overview of various ML, evolutionary algorithms, and soft computing techniques surveyed and analysed for de novo drug design, emphasising the computational aspects. Copyright © 2022 Inderscience Enterprises Ltd.