3D QSAR STUDIES, VIRTUAL SCREENING AND MACHINE LEARNING OF NOVEL PROTEIN KINASE C DERIVATIVES TO OBTAIN NEW INHIBITORS FOR CANCER

Abstract

Protein kinase C (PKC) family is a group of kinases that have always been the focus of pharmaceutical industries for the last two decades. These kinases are capable of modulating important cellular functions such as differentiation, proliferation and even cell survival. The over expression of novel class of PKC has been reported in many benign cancers which further leads to malignant if left unchecked. Therefore inhibiting n-PKC by effective compounds is necessity. In this study 3D-QSAR modeling was performed on a series of novel class of Protein Kinase C derivatives acting as inhibitors in various cancers. The compounds were collected from two datasets with the same scaffold, and utilized as a template for a new model to screen the ZINC database and Binding database of commercially available derivatives. The datasets were divided into training and test sets. As the first step, comparative analysis was conducted out by Swiss ADME and top features were selected to create similar inhibitors by Marvin JS package and molecular property and bioactivity scores were calculated. The constructed compounds were used as test set in machine learning to create various models. In parallel docking studies were applied to a set of known n-PKC inhibitors and constructed inhibitors. The validity and the prediction capacity of the resulting models were further evaluated. It is crucial for developing targeted therapy by use of specific inhibitors that have more rapid action than current available treatments.