Insilico modelling of quantitative structure–activity relationship of pGI50 anticancer compounds on K-562 cell line

Abstract

Abstract The pGI50 cytotoxicity values of 112 compounds on K-562 cancer cell line were modelled in order to illustrate the quantitative structure–activity relationship of the compounds. The data set were divided into training and test set through Kennard-stone algorithm, while the pool of molecular descriptors calculated with paDEL descriptor metric program was subjected to genetic functional algorithm for selection of descriptor to be modeled. The statistical significance of the model was verified by calculating the values of Q2LOO (0.845), Q2F1 (0.9397), Q2F2 (0.6862) and R2pred (0.6862) needed to evaluate the strength and robustness of the model. The result of the internal and external validation of the model indicates that the model is good and could be used to predict the GI50 of anticancer compounds on K-562 leukemia cell line.