Abstract
Antagonism of cannabinoid receptor-1 has emerged as a most promising therapeutic target for the development of anti-obesity drugs. In the present study, an in silico approach using decision tree, random forest and moving average analysis has been applied to a data set comprising of 76 analogues of substituted 2-(3-pyrazolyl)-1,3,4-oxadiazoles for development of models for prediction of antagonistic activity of cannabinoid receptor-1. A total of 46 2D and 3D molecular descriptors of diverse nature were employed for decision tree and random forest analysis. The values of majority of these descriptors for each analogue involved in the dataset were computed using E-Dragon software (version 1.0). Random forest correctly classified the analogues into active and inactive with an accuracy of 95%. A decision tree was also utilized for determining the importance of molecular descriptors. The decision tree learned the information from the input data with an accuracy of 99% and correctly predicted the cross-validated (10 fold) data with an accuracy up to 90%. Finally, three molecular descriptors of diverse nature (including best descriptor identified by decision tree analysis) were subsequently used to build suitable models using moving average analysis. These models resulted in the prediction of cannabinoid receptor-1 antagonistic activity with an accuracy of 95–96%. High predictability of proposed models offer vast potential for providing lead structures for the development of potent cannabinoid receptor-1 antagonists for the treatment of obesity.