Quantitative structure activity relationship (QSAR) studies have been performed on the 27 molecules belonging
to a series of 2,3,6-trisubstituted quinazolin-4(3H)-one derivatives for their antitumor activity. To explore the relationship
between structure and activity, various chemometric tools have been employed such as Factor Analysis-Multiple Linear
Regressions (FA-MLR), Artificial Neural Network (ANN) analysis and Principle Component Regression (PCR) methods.
The models exhibited good correlation coefficient (r2) and cross validated correlation coefficient (q2) for all methods. It
was found that ANN method gave better results indicating that the topological (IC4 and MPC06), constitutional (nf) and
geometrical (G (N..S)) parameters were the most significant parameters for the antitumor activity indicating that novel potent
analogs can be synthesized by altering these descriptor characteristics.
Keywords: QSAR, antitumor activity, FA-MLR, ANN, PCR.
Rights & PermissionsPrintExport