In silico modeling of some HEPT analogues as anti -HIV agents using QSAR and molecular docking studies

Quantitative structure–activity relationship (QSAR), Density functional theory and molecular docking studies have been performed on a series of 49 1-[2-Hydroxyethoxy) methyl]-6-(phenylthio)-thymine (HEPT) derivatives as anti-HIV agents. A genetic algorithm multiple linear regression (GA-MLR) analysis has revealed that the five variable model comprises of 2D autocorrelation, 2D atom pairs and constitutional descriptors of these compounds are the governing factors for their biological activity. Using this GA-MLR model some new compounds have been proposed that have higher potency than the existing ones. The molecular docking study has shown that these new compounds can form hydrogen bonds with the receptor and have effective steric interactions.