Molecular docking, 2D and 3D-QSAR studies of new indole-based derivatives as HCV-NS5B polymerase inhibitors

Abstract

Hepatitis C virus (HCV) is a main cause of liver disease worldwide and a potential cause of considerable malady and mortality in the future. The docking and molecular dynamics simulation (MDs) have been carried out on a series of indole-based derivatives into the cavity of HCV-NS5B polymerase. The docking and MDs led to recognition of the best conformers of inhibitors and also determination of the key interacting amino acids of the binding site of the protein packet. In addition, the conformers are very similar to bioactive ones which in turn will guarantee the most reliable and predictive models in 2D and 3D-QSAR studies. To find the correlation between inhibitory activities of ligands against HCV-NS5B, a comparative molecular field analysis (CoMFA) study, as a 3D-QSAR method, was carried out and the corresponding contour maps of electrostatic and steric fields were computed. Furthermore, 2D descriptors were calculated utilizing the optimal conformers. In the case of 2D-QSAR, a variable selection was applied by genetic algorithm (GA), followed by model building using partial least squares (GA-PLS) and support vector machine (GA-SVM) regression methods. The predictive ability of the proposed models was validated by a structurally diverse test set of 22 compounds that had not been included in the training step. The q ² and the $r_{\text{pred}}^{2}$ values for CoMFA, GA-PLS and GA-SVM models were 0.521, 0.655, and 0.746 and 0.659, 0.881, and 0.967, respectively. Moreover, the molecular interactions of these inhibitors with the HCV-NS5B’s active site residues were properly discussed.