Using associative classification for predicting HIV-1 drug resistance

Drug resistance testing, genotyping or phenotyping, is an important role in management of HIV-1 infections. To overcome the drawbacks of genotyping and phenotyping in predicting the HIV-1 drug resistance, predicting of phenotypic resistance from genotypic data is an interesting task. In this paper, the CBA algorithm was used to discover the relationship between the amino acid positions and drug susceptibility and to construct the classifiers to predict phenotypic resistance for 6 protease inhibitors. The performance of the prediction was measured by 10-fold cross-validation. The best model provided the accuracy between 84.11% and 92.64% for all 6 protease inhibitors. In addition, the average accuracy of 6 drugs of the prediction using the CBA algorithm provided the best performance when compared with HIVdb, SVM, and REG algorithms.