Application of GA-MLR for QSAR modeling of the arylthioindole class of Colchicine binding inhibitors as anticancer agents

The GA-MLR is a powerful search technique based on the evolution of biological systems for QSAR modeling. In this study we had study QSAR modeling of some Arylthioindoles which obtained by replacing the 2-alkoxycarbonyl group with a bioisosteric five- membered heterocycle nucleus. The general Hypothesis is that these derivatives can be useful for designing and synthesizing anti-cancer drugs, since they are a group of potent inhibitors of Colchicine binding and cancer cell growth. For external validation of QSAR model, the data set was split into the training and test sets by using random splitting method. The random sampling of the training set (80% of data) was performed 20 times and the remaining molecules were used as external validation set. The GA-MLR method as variable selection method was applied on 20 random training data set. The best multivariate linear model based on and values had five parameters GA-MLR method and the value was 0.6209 for GA-MLR method. The results were indicated that GA-MLR method is a powerful method for variable selection by comparison with other methods.

Keywords:

Cancer , GA-MLR , Arylthioindoles , Colchicine , random splitting