Performance of genotypic tools for prediction of tropism in HIV-1 subtype C V3 loop sequences

Abstract

Currently, there is no consensus on the genotypic tools to be used for tropism analysis in HIV-1 subtype C strains. Thus, the aim of the study was to evaluate the performance of the different V3 loop-based genotypic algorithms available. We compiled a dataset of 645 HIV-1 subtype C V3 loop sequences of known coreceptor phenotypes (531 R5-tropic/non-syncytium-inducing and 114 X4-tropic/R5X4-tropic/syncytium-inducing sequences) from the Los Alamos database (http://www.hiv.lanl.gov/) and previously published literature. Coreceptor usage was predicted based on this dataset using different software-based machine-learning algorithms as well as simple classical rules. All the sophisticated machine-learning methods showed a good concordance of above 85%. Geno2Pheno (false-positive rate cutoff of 5-15%) and CoRSeqV3-C were found to have a high predicting capability in determining both HIV-1 subtype C X4-tropic and R5-tropic strains. The current sophisticated genotypic tropism tools based on V3 loop perform well for tropism prediction in HIV-1 subtype C strains and can be used in clinical settings.