Abstarct: Recent investigations show that estrogen plays important role in breast cancer disease. Thus some researches have been carried out on compounds which can inhibit pathways of this hormone in body at some ways. These studies are focused on the two categories of compounds. The first group of compounds directly influence on estrogen activities. The second group of compounds inhibits the estrogen production procedures. One group of these compounds that can inhibit estrogen production is Sufonanilide derivatives. In the first section of this thesis, QSAR models for 21 analogues of Sulfonanilide were constructed using MLR and Bayesian regularized ANN (BR-ANN). Among of a large number of calculated descxriptors only 20 significant molecular descxriptors were obtained by stepwise regression, as the most feasible descxriptors, and then 6 descxriptors were used as inputs for ANN. The data set was randomly divided into train (16 molecules) and test (5 molecules) sets and the models parameters and ANN architecture were optimized by cross-validation method. The prediction ability of each model was evaluated using the test data set and Leave-One-Out cross-validation (LOO) Method. The MSEs for the test data set and LOO method were 0.0227 and 0.0132 for MLR, respectively. The MSEs values obtained from applying ANN model to test set and cross-validation were 0.0098 and 0.0161, respectively. The results obtained from ANN showed the excellent prediction of the inhibitory activity data of the corresponding analogues. In the second section, quantitative-structure activity of some 5-oxopyrrolidine-3-carboxamide derivatives was studied. These compounds suppress disease by interfering in interaction between CCR5 and HIV. Data set consist of 104 analogues of 5-oxopyrrolidine-3-carboxamide. In this study, data set was divided into 62 trains, 21 validations and 21 test sets. The ANN model for prediction of pIC50 was constructed using 19 descxriptors. MSE values for test and LOO method were 0.1210 and 0.1405, respectively.