Comparative (Q)SAR analysis of benzodiazepine derivatives with different biological activity.

211 compounds containing a benzodiazepine moiety (BZD) and belonging to 4 groups of different biological activity (H - inhibitors of reverse transcriptase of HIV-I virus, A - antiarrhythmic agents, G - ligands of benzodiazepine receptor in GABAergic system and C - cholecystokinin receptor antagonists) were subjected to structure-activity relationship (SAR) analysis. SAR investigations of all 211 BZD were based on Discriminant Function Analysis (DFA) of physicochemical data connected with BBB (blood-brain barrier) permeability of studied compounds. DFA was performed with STATISTICA 10.0 software by the stepwise method and resulted in 3 discriminant functions whose quality was assessed by Wilk's lambda parameter. Calculated discriminant functions (roots) were applied to draw the scatter diagram of canonical values that showed all 211 cases divided into 4 groups of different biological activity. The method was successfully validated with a set of 38 BZD derivatives expected to belong to groups H, A, G and C. The reliability of the obtained model was confirmed with a cross-validation test. Classification functions presented in this study may be used as a practical tool for predicting new BZD drugs activity.