Estimating HIV-1 Viremic Time From Reservoir Sequence Diversity With Uncertainty Quantification.

Area of Science:

• Virology
• Immunology
• Computational Biology

Background:

• Accurate estimation of human immunodeficiency virus (HIV) viremic time is crucial for understanding reservoir dynamics and guiding cure trials.
• Traditional methods using serological assays or CD4 counts lack quantitative precision.
• Sequence-based estimates are limited by the increasing use of immediate antiretroviral therapy (ART) initiation.

Purpose of the Study:

• To develop and validate Bayesian models for predicting HIV viremic time using sequence diversity from HIV reservoir sequences.
• To evaluate the performance of different diversity metrics and modeling strategies.

Main Methods:

• Developed Bayesian models to predict viremic time using six diversity metrics from gp41, RT, and matrix p17 regions in HIV reservoir sequences.
• Fitted 36 Bayesian models per region using slope-fitting and weighting strategies.
• Validated models on participants with known diagnosis dates and evaluated predictive accuracy and model diagnostics.

Main Results:

• Reservoir sequence diversity positively correlated with viremic time across all metrics.
• Models using unique RT and gp41 sequences, particularly with simple diversity metrics (nucleotide diversity, mean TN93 distances), showed strong predictive accuracy.
• Validation demonstrated that models produced estimates aligning with known HIV diagnosis dates, with improved precision using log-transformed sequence counts.

Conclusions:

• A novel Bayesian approach effectively estimates HIV viremic time from reservoir sequences, providing uncertainty estimates.
• The method is applicable across HIV subtypes and chronic infections, utilizing simple diversity metrics.
• This approach can significantly support research into HIV reservoir dynamics and the development of HIV cure strategies.