Background: Apoptosis is known as programmed cell death that plays an important role in tumor biology.
Methods: In this study, apoptosis-inducing activity is predicted by using a QSAR modeling approach for a series of 4-anilinoquinozaline derivatives. 2D-QSAR model for the prediction of apoptosis-inducing activity was obtained by applying multiple linear regression giving r2 = 0.8225 and q2 = 0.7626, principal component regression giving r2 = 0.7539 and q2 = 0.6669 and partial least squares giving r2 = 0.8237 and q2 = 0.6224.
Results: QSAR study revealed that alignment-independent descriptors and distance-based topology index are the most important descriptors in predicting apoptosis-inducing activity. 3D-QSAR study was performed using k-nearest neighbor molecular field analysis (kNN-MFA) approach for both electrostatic and steric fields. Three different kNN-MFA 3D-QSAR methods (SW-FB, SA, and GA) were used for the development of models and tested successfully for internal (q2 > 0.62) and external (predictive r2 > 0.52) validation criteria. Thus, 3D-QSAR models showed that electrostatic effects dominantly determine the binding affinities.
Conclusions: The QSAR models developed in this study would be useful for the development of new apoptosis inducer as anticancer agents.