MIA-QSAR modelling of anti-HIV-1 activities of some 2-amino-6-arylsulfonylbenzonitriles and their thio and sulfinyl congeners

Abstract

A QSAR method based on MIA (multivariate image analysis) descriptors is applied to a series of 2-amino-6-arylsulfonylbenzonitriles and their thio and sulfinyl congeners, compounds with anti-HIV-1 activity. Two models were built in order to appraise the modelling capability when different drawing programs are used to create the set of molecules. Both models showed good predictive ability, with cross-validated Q2 of 0.712 and 0.624, and Q2 for an external validation set of 0.823 and 0.747. An ADME evaluation, by calculating the topological polar surface area (TPSA) and parameters derived from the rule of five, was also performed to proposed compounds in order to suggest absorption profiles for potential new drugs.