Predicting drug resistance of the HIV-1 protease using molecular interaction energy components.

ABSTRACT: Drug resistance significantly impairs the efficacy of AIDS therapy. Therefore, precise prediction of resistant viral mutants is particularly useful for developing effective drugs and designing therapeutic regimen. In this study, we applied a structure-based computational approach to predict mutants of the HIV-1 protease resistant to the seven FDA approved drugs. We analyzed the energetic pattern of the protease-drug interaction by calculating the molecular interaction energy components (MIECs) between the drug and the protease residues. Support vector machines (SVMs) were trained on MIECs to classify protease mutants into resistant and nonresistant categories. The high prediction accuracies for the test sets of cross-validations suggested that the MIECs successfully characterized the inter

SUBMITTER: Hou T 

PROVIDER: S-EPMC3210569 | biostudies-literature | 2009 Mar

REPOSITORIES: biostudies-literature