Modeling of novel HIV-1 protease inhibitors incorporating N-Aryl-oxazolidinone-5-carboxamides as P2 ligands using quantum chemical and topological finger print descriptors

This article presents two QSAR studies on the novel series of HIV-1 protease inhibitors incorporating N-phenyloxazolidinone-5-carboxamides into the (hydroxyethylamino) sulfonamide scaffold as P2 ligands. All the statistical calculations have been done using PRECLAV software. A heuristic algorithm selects the best multiple linear regression (MLR) equation according to the highest leave-one-out cross-validation correlation coefficient. In the first QSAR study, only the Quantum Chemical descriptors have been used. The correlation between the observed values and the calculated values of activity is good (r (2) = 0.935, Se = 0.3782, r (2)cv = 0.900, F = 76.6568). The second QSAR study has used both quantum chemical descriptors and topological finger print descriptors. The correlation between the observed values and the calculated values of activity is very good (r (2) = 0.9638, Se = 0.2823, r (2)cv = 0.9492, F = 141.8338). The virtual molecular fragments that lead to a significant increase of the inhibitor activity are CH3 and CH. The virtual fragments C4H3S and C6H2 lead to significant decreases in the inhibitory activity. The results are critically discussed using Ridge statistics.