QSAR studies and molecular docking analysis of 1,4-dihydropyridines as Calcium Channel Blockers using DFT and ANN methods

Abstract Artificial neural networks (ANNs) are useful for predicting biological activities from large datasets of molecules. Unlike traditional statistical methods such as regression analysis, ANNs allow the study of complex and nonlinear relationships such as QSAR studies. Here, we use artificial neural network and multiple linear regression (MLR) methods to generate QSAR models for Calcium Channel Blockers activity of a series of 1,4-dihydropyridine derivatives molecules. The molecular descriptors were calculated by using Density Functional Theory (DFT) method at the B3LYP/6-31G+ (d, p) level. The statistical analyses indicate that the predicted values are in good agreement with the experimental results for both the training and test sets using either MLR or ANN. In addition, we used molecular docking to determine the binding energies, and ligand-protein interactions between these compounds and their biological target.