Chapter 9 - Advances in QSAR through artificial intelligence and machine learning methods

Abstract

The goal of drug discovery is to locate lead compounds and novel compounds that have certain features and potential to be developed into drugs to treat diseases. The application of machine learning (ML), a branch of artificial intelligence (AI), has led to its widespread adoption as an essential component in modern computer applications in recent decades. The establishment of fresh models requires the development of new ML and AI algorithms. These tools are particularly useful in the area of quantitative structure–activity relationships (QSARs). One ML algorithm, random forest, was recently introduced to develop QSAR for predicting biological activity of compounds based on the description of the compounds' molecular structures. It also provides a variable that is of crucial importance to the measurement of the reduction wrapper algorithm. In this chapter, we focus on the various AI and ML algorithms that have been used in recent years of research, specifically on QSAR, to analyze the state of the art in the drug development process. These algorithms have also been used to model the molecular data and biological problems that are used in the drug discovery process. In addition, we have included in this chapter an overview of the fundamentals of QSAR in drug discovery as well as a discussion of the research tools that are now accessible.