Computational approach for generating robust models for discovering novel molecules as Cyclin Dependent Kinase 4 inhibitors

Abstract

Cyclin Dependent Kinase 4 is a striking target for the proposal of anti-cancer drugs since its overexpression is associated with various types of cancers. In the present study, 2D and 3D atom based QSAR study were accomplished with 6 component PLS factor for 230 pyrido[2,3-d]pyramidine correspondents along with flexible ligand docking in the extra precision mode with the application of core constraints followed by the binding energy determinations. Kernel based partial least square analysis fitting with fingerprints initially created worthy models, among which the one with molprint2D fingerprints generated a noble model with a score value of 0.8322. Atom Based 3D QSAR resulted in an effective model with Regression coefficient (R2 = 0.8372), and Q2 = 0.7381. Docking experiments exposed hydrogen bonding interactions with hinge region residues, salt bridge formation and л-л stacking interaction as the leading non-covalent interactions causative of the inhibitory activity of CDK4 inhibitors. The primary factors that induced the stability of protein-ligand complex are the van der Waals interactions, lipophilic interactions and coulombic interactions.