Molecular clock dating using complex mixture models: applied to ancient symbionts.

Area of Science:

• Evolutionary Biology
• Computational Biology
• Phylogenetics

Background:

• Molecular clocks are crucial for evolutionary timing but often inaccurate due to simple substitution models.
• Substitution heterogeneity and saturation in deep-time or fast-evolving lineages compromise current molecular clock methods.

Purpose of the Study:

• Introduce phyloHessian, a Julia-based software for molecular dating using complex mixture substitution models.
• Enhance the accuracy of divergence time and substitution rate estimates in molecular clock analyses.

Main Methods:

• phyloHessian computes the phylogenetic Hessian matrix.
• Integrates the Hessian matrix into PAML-MCMCtree's approximate likelihood framework for dating.
• Utilizes complex mixture substitution models instead of homogeneous ones.

Main Results:

• Complex mixture models significantly improve divergence time and substitution rate accuracy in deep-time phylogenies.
• Mixture models show greater robustness to model and calibration uncertainties compared to homogeneous models.
• Empirical analyses revealed accelerated substitution rates and altered divergence times for Microsporidia and Rickettsiales.

Conclusions:

• Incorporating complex mixture substitution models is vital for reliable evolutionary timelines.
• phyloHessian enhances the study of deep-time or fast-evolving lineages.
• Revised evolutionary histories and host association origins are suggested for ancient symbionts.