Abstract
In this study, a nonlinear quantitative structure-activity relationship model for the prediction of the IC50 of trisubstituted thiazoles as Cdc7 kinase inhibitors was developed by the gene expression programming (GEP). This model is based on five descriptors which were selected from the descriptors’ pool by heuristic method (HM). GEP rendered a good correlation between the experimental and predicted log (IC50) values with a correlation coefficient and a mean error of 0.85 and 0.22 for the training set, 0.84 and 0.39 for the testing set, respectively. The results indicate that this QSAR model has good predictive capability and it points to further optimization opportunities for Cdc7 kinase inhibitors.
Keywords: Gene expression programming, IC50, kinase inhibitors, quantitative structure activity relationship, cell division cycle 7, trisubstituted thiazoles.
Graphical Abstract