Abstarct: In the first section, quantitative structure-activity relationship (QSAR) study was conducted on the inhibition effect (pIC50) of 103 drug compounds of homopiperazine, diamine, 3-aminopyrrolidine derivatives.Two variable selection methods of stepwise regression (SR) and genetic algorithm (GA) were applied to select the important descxriptors. The three methods of multiple linear regression (MLR), artificial neural network (ANN) and support vector machines (SVM) were used for construction of methods with selected descxriptors for prediction of pIC50 of these compounds. The performance of each model was investigated by the test set. The mean square error (MSE) and standard error of prediction (SEP) for the test sets of SR-MLR, SR-ANN, SR-SVM, GA-MLR, GA-ANN and GA-SVM were 0.328 , 0.573 and 0.346 , 0.588 and 0.276 , 0.513 and 0.228 , 0.477 and 0.225 , 0.475 and 0.257, 0.532 respectively. In the second section, quantitative structure-activity relationship (QSAR) study was conducted on the inhibition effect (pIC50) of 42 drug compounds of Arylsulfonylpiperazine derivatives. The best calculated descxriptors were selected by genetic algorithm. The multiple linear regression (MLR), artificial neural network (ANN) and support vector machines (SVM) were applied construction of models to predict of the inhibition effect (pIC50) of these compounds. The performance of each model was investigated by test set. The mean squared error (MSE) and standard error of prediction (SEP) and the coefficient of determination (R2) very close to each other, and this reflects to this fact, that all three models have good ability to predict the pIC50 of the studied compounds.