QSAR and anticancer drug design of β-Carboline compounds utilizing computed molecular descriptors
Ruby Yadav, N Sisir · Journal of computational methods in molecular design · 2014
An attempt has been made for the development of qua ntitative structure-activity relationship (QSAR) mo dels for a series of �-carboline derivatives having DYRK1A (dual specific ity tyrosine phosphorylated and regulated kinase 1a ) inhibitory activities as potent anticancer agents t oward the activation of caspase-9 which leads to ma ssive apoptosis in different human cancer cell types including glio ma, esophageal cancer and non-small-cell lung cance rs respectively. A number of highly descriptive and pr edictive QSAR models for these compounds were obtai ned by considering in vitro anticancer activities against glioma cell lines including U373 and Hs683 respecti vely with various sets of theoretical molecular descriptors i ncluding topological, constitutional, geometrical, functional groups and atom centered fragment indices calculate d solely from the structures of 48 synthesized �-carboline derivatives using stepwise-multiple linear regressi ons methods. Model validation is performed by incor porating training and test sets approach and calculating R 2 , Q Loo 2 , R pred 2 and standard error of estimation (SEE) respectivel y. From these models a number of significant features of these congeners including X3Av, EEig13x, MATS5m and SP05 which are responsible for size, shape and weig ht of the molecules whereas EEig03d, MATS4p and QYY p which indicate dipole moment, polarizabilities, con jugation and aromaticity have been predicted for th e design of more promising �-carboline anticancer compounds.