The search and development of anti-HIV drugs is currently one of the most urgent tasks of pharmacological studies. In this work, a quantitative structure-activity relationship (QSAR) model based on some new norm indexes, was obtained to a series of more than 150 HEPT derivatives (1-[(2-hydroxyethoxy) methyl]-6-(phenylthio) thymine) to find their pEC(50) (the required effective concentration to achieve 50% protection of MT-4 cells against the cytopathic effect of virus) and pCC(50) (the required cytotoxic concentration to reduce visibility of 50% mock infected cell) activities. The model efficiencies were then validated using the leave-one-out cross validation (LOO-CV) and y-randomization test. Results indicated that this new model was efficient and could provide satisfactory results for prediction of pEC(50) and pCC(50) with the higher R-train(2) and the higher R-test(2). By using the leverage approach, the applicability domain of this model was further investigated and no response outlier was detected for HEPT derivatives involved in this work. Comparison results with reference methods demonstrated that this new method could result in significant improvements for predicting pEC(50) and pCC(50) of anti-HIV HEPT derivatives. Moreover, results shown in this present study suggested that these two absolutely different activities pEC(50) and pCC(50) of anti-HIV HEPT derivatives could be predicted well with a totally similar QSAR model, which indicated that this model might have the potential to be further utilized for other biological activities of HEPT derivatives. (C) 2016 The Chemical Industry and Engineering Society of China, and Chemical Industry Press. All rights reserved.