Bayesian Phylogenetic Inference using Relaxed-clocks and the Multispecies Coalescent.

Area of Science:

• Computational Biology
• Phylogenetics
• Evolutionary Genomics

Background:

• The multispecies coalescent (MSC) model is a robust framework for analyzing genomic data, particularly when gene trees vary across the genome.
• Existing MSC implementations often assume a strict molecular clock and simple substitution models, limiting their applicability to closely related species.

Purpose of the Study:

• To extend the MSC model to incorporate more general substitution models and relaxed molecular clocks.
• To enable accurate estimation of species divergence times and ancestral population sizes when the strict clock assumption is violated.
• To provide a simulation framework for evaluating species tree estimation methods under relaxed clock conditions.

Main Methods:

• Implementation of general time-reversible (GTR) substitution models within the MSC framework.
• Development and integration of relaxed molecular clock models allowing rate variation among species.
• Utilizing Markov chain Monte Carlo (MCMC) algorithms for phylogenetic inference.

Main Results:

• The extended MSC model with relaxed clocks accurately estimates phylogenetic parameters even when molecular clock assumptions are significantly violated.
• The clock-JC model remains adequate for closely related species, but relaxed clocks are crucial for more distantly related ones.
• Phylogenetic information in gene-tree branch lengths can be effectively extracted by relaxed-clock models.

Conclusions:

• The new MSC models with relaxed clocks offer significant improvements for estimating species divergence times and population parameters, especially for deeper evolutionary timescales.
• While MCMC algorithms show promise, mixing issues require further investigation and potential improvements for robust species tree estimation under relaxed clocks.