The drug discovery method was revolutionized by AI, which may rapidly select physiologically active chemicals from millions of candidates. This article summarizes several applications of ML-based tools, such as GOLD, Deep PVP, LIB SVM, etc., and the associated methodologies, like SVM, RF, decision tree, and ANN, etc., at numerous phases of drug creation & development. They can utilize these methods for SNP discovery, drug repurposing, ligand-based drug design, Ligand-based Virtual Screening, Structure-based Virtual Screening, Leading identifications, quantitative structure-activity relationship modelling, & ADMET evaluation. Human intestinal absorption predictions would benefit significantly from the superior accuracy displayed by SVM. Cases have been published that illustrate the effectiveness of SVM & RF models in finding JFD00950 as a new drug targeting a colon cancer cell line, DLD-1, by inhibiting the cytotoxic & cleavage activities of FEN1. They also utilized a QSAR approach to estimate flavonoid inhibitory effects on AR activity as a potent therapy for diabetic Mellitus (DM) using ANN. Consequently, ML techniques have developed as reliable in the era of big data. A practical tool to cope with the vast volumes of information created by current drug development to model small-molecule medicines and gene biomarkers, & uncover novel drug targets for various diseases.

ML, AI, drug discovery.