Conformation depends on 4D-QSAR analysis using EC-GA method: pharmacophore identification and bioactivity prediction...

The electron conformational and genetic algorithm methods (EC-GA) were integrated for the identification of the pharmacophore group and predicting the anti HIV-1 activity of tetrahydroimidazo[4,5,1-jk][1,4]benzodiazepinone (TIBO) derivatives. To reveal the pharmacophore group, each conformation of all compounds was arranged by electron conformational matrices of congruity. Multiple comparisons of these matrices, within given tolerances for high active and low active TIBO derivatives, allow the identification of the pharmacophore group that refers to the electron conformational submatrix of activity. The effects of conformations, internal and external validation were investigated by four different models based on an ensemble of conformers and a single conformer, both with and without a test set. Model 1 using an ensemble of conformers for the training (39 compounds) and test sets (13 compounds), obtained by the optimum seven parameters, gave satisfactory results (R²(training) = 0.878, R²(test)= 0.910, q² = 0.840, q²(ext1) = 0.926 and q²(ext2) = 0.900).