Prediction of pKa Values of Benzimidazoles Using Genetic Algorithm

Benzimidazoles and its derivatives are a large chemical and pharmaceutical family used to treat vermicides or fungicides as they inhibit the action of certain micro organisms. In the present work, pka values for set of 25 benzimidazoles derivatives has been investigated by genetic algorithm (GA). Semi-empirical quantum chemical calculations at AM1 level were used to find the optimum 3D geometry of the studied molecules. Then different quantum-chemical descriptors were calculated by HyperChem and Dragon softwares and after optimizing of the GA parameters, the best descriptors were selected by GA, and then these descriptors were applied as input in a multi linear regression model for prediction of pka for benzimidazoles derivatives. GA gave good statistical results both in calibration (R2 = 0.9855) and prediction (R2 = 0.9715) series. The results show that applied procedure was suitable for prediction of pka for benzimidazoles derivatives.