In this research work Genetic function algorithm was employed to generate the QSAR models due to its ability to produce a vast population of model instead of just a single model. Four models were generated from the model building set and the first one was chosen because of its statistical significance.
The experimental activities of the datasets were normalized to pIC50. Also, the predicted pIC50(s) of the compounds was determined (Table 1). The differences between the experimental and predicted activities of the compounds (residual values) were further calculated (Table 1).