A reliable phenotype predictor for human immunodeficiency virus type 1 subtype C based on envelope V3 sequences.

Area of Science:

• Virology
• Bioinformatics
• Genetics

Background:

• Human immunodeficiency virus type 1 (HIV-1) subtype C causes over 50% of global infections.
• CXCR4 coreceptor usage in HIV-1 subtype C is less understood than in subtype B, despite its association with disease progression.
• Accurate prediction of CXCR4 variants is crucial for understanding HIV-1 subtype C pathogenesis.

Purpose of the Study:

• To develop and validate a genotypic prediction method for CXCR4 coreceptor usage in HIV-1 subtype C.
• To compare the performance of the new method against existing prediction strategies.

Main Methods:

• A Position-Specific Scoring Matrix (PSSM) approach, previously successful for subtype B, was adapted for subtype C V3 loop sequences.
• A training set of 279 subtype C sequences with known phenotypes (CCR5+ NSI and CXCR4+ SI) was used to derive the C-PSSM predictor.
• Performance was evaluated using bootstrapping and leave-one-out cross-validation, with a separate validation set.

Main Results:

• The C-PSSM predictor achieved a specificity of 94% and a sensitivity of 75% on the training data.
• The method demonstrated significantly higher sensitivity (75%) compared to other methods (e.g., charged residue method at 47.8%).
• Validation on a unique subtype C set yielded a specificity of 83% and sensitivity of 83%.

Conclusions:

• The developed C-PSSM predictor is a reliable and sensitive tool for predicting CXCR4 coreceptor usage in HIV-1 subtype C.
• This bioinformatic approach offers a rapid and cost-effective means to identify CXCR4 variants, aiding research into subtype C progression.
• The findings suggest that specific genetic sites influencing coreceptor usage may differ between HIV-1 subtypes.