Abstarct: Non-nucleoside reverse transcxriptase inhibitors (NNRTIs) are a class of anti retroviral drugs used to treat HIV infection. NNRRTIs inhibit activity of reverse transcxriptase (RT), an enzyme that controls the replications of the genetic material of HIV. In the first section, Quantitative structure – activity relationship (QSAR) models were constructed for predicting the Anti-HIV activity of a set of Thiocarbamate Derivatives with group contribution descxriptors and Random forest method. Data set was randomly divided into training set and test set including 157 and 40 compounds, respectively. The validation study of the RF models was performed using test set and leave-one-out technique. The results obtained for prediction of anti HIV activity of the test set and leave-one-out method by RF models showed squared correlation coefficients of 0.8748 and 0.8025, respectively. In the second section, some derivatives of necroptosis inhibitors (Necrostatin-5) were studied. In this study, A data set including 160 compounds were used to design predictive models. A new method Random Forest was used for classification necrostatin-5 compounds in to active and inactive groups. For a series of the 51 active componds, The most effective descxriptors were selected using Random Forest and Rtepwise Regression(RF-SR) method. The selected descxriptors were correlated with the bioactivities of the molecules using the well known Artificial Neural Networks(ANN). The prediction ability of the proposed models was evaluated by test set and leave-one-out method. The squared correlation coefficients obtained for test set and leave-one-out method by ANN models were 0.9677 and 0.8295, respectively.