Prediction of Drug Resistance Mechanisms: Integration of Molecular Modeling and Artificial Intelligence Methods

Abstract Drug resistance caused by single-point amino acid substitutions in target proteins represents a serious problem in modern therapy of oncological, infectious and viral diseases. This review systematizes modern computational modeling methods that allow studying the molecular mechanisms of such resistance and predicting its emergence. Contemporary databases and machine learning tools used for resistance prediction from genomic data are analyzed. The potential of the emerging direction of physics-informed artificial intelligence models for enhancing the interpretability of predictions is noted.