Divergence dating using mixed effects clock modelling: An application to HIV-1.

Area of Science:

• Evolutionary biology
• Computational biology
• Virology

Background:

• Estimating divergence times is crucial for evolutionary studies.
• Relaxed molecular clock models are widely used but can be biased by rate variation.
• Existing models struggle with discrete rate variation among lineages, as seen in HIV-1.

Purpose of the Study:

• To investigate the utility of mixed effects molecular clock models for estimating divergence times.
• To assess the performance of mixed effects models compared to existing methods.
• To analyze rate variation in HIV-1 group M evolution.

Main Methods:

• Simulations were used to compare model performance.
• Mixed effects molecular clock models were applied to a comprehensive HIV-1 group M genome dataset.
• Bayesian phylogenetic inference was employed.

Main Results:

• Mixed effects models outperform existing models in simulations with mixed rate variation.
• Significant rate variation among HIV-1 subtypes was confirmed, not adequately captured by uncorrelated clock models.
• The mixed effects model estimated the HIV-1 group M common ancestor at 1920 (1915-25).

Conclusions:

• Mixed effects molecular clock models offer a robust solution for estimating divergence times with complex rate variation.
• Complete genome data significantly reduces uncertainty in divergence time estimates compared to shorter gene sequences.
• This approach enhances the accuracy of viral evolutionary history reconstruction.