AI-Assisted Methods for Protein Structure Prediction and Analysis

Proteins are the workhorses of cells. Their sequence is determined by the genetic code embedded in the DNA, which translates it faithfully into a string of amino acids known as the primary structure of proteins. But for proteins to achieve functional mode, they must be correctly folded into a three-dimensional structure commonly known as their tertiary structure. Determining the tertiary structure of the proteins is often an expensive and time-consuming process. Protein structure prediction has been in play for several decades now. But recent developments in the fields of computational hardware, software, and artificial intelligence have led to the simultaneous development of methods capable of protein secondary and tertiary structure prediction. Machine learning-based methods have recently emerged as aids of choice for the prediction of protein structures. Programs like AlphaFold and AlphaFold2 have revolutionized the structure prediction landscape. This chapter presents a comprehensive account of the basics of AI-associated methods, along with the evolution of protein structure prediction and its subsequent analysis, helping in related applications in diverse fields ranging from drug discovery to enzyme design.