QSAR and Drug-likeness Studies of Thiadiazole Derivatives Against Lung Cancer

Abstract

This study was aimed at building a robust quantitative structure–activity relationship (QSAR) to predict the anti-proliferate activity of 1,3,4-thiadiazole derivatives against the A549 lung cancer cell lines. The semi-empirical PM7 parametrization approach was used to optimize the complete set of 1,3,4-thiadiazole derivatives and various classes of molecular descriptors have been calculated. We built models using Fisher score and the best subset selection for feature selection, and the final model was developed using the multiple linear regression technique, all in accordance with the rigorous Organization for Economic Co-operation and Development (OECD) requirements. Furthermore, various internationally agreed severe validation parameters were used to validate the model. Overall, our established model for quick prediction should be relevant to new, untested, or not yet produced compounds that fall within the applicability domain (AD) of the model. The drug-likeness properties of the 10 compounds with the greatest activity value were also calculated using Lipinski's rule properties. Keywords: QSAR, Thiadiazole derivatives, A549, PM7, OECD