Bayesian Estimation of Species Divergence Times Using Correlated Quantitative Characters

Area of Science:

• Evolutionary Biology
• Phylogenetics
• Quantitative Morphology

Background:

• Discrete morphological data are common in evolutionary studies, but quantitative (continuous) data are less utilized.
• Accurate estimation of species divergence times is crucial for understanding evolutionary history.
• Existing methods may not fully account for complexities in quantitative character evolution.

Purpose of the Study:

• To implement a novel Bayesian method for estimating species divergence times using quantitative morphological characters.
• To assess the impact of character correlation and population variation on divergence time and rate estimates.
• To integrate quantitative morphological data with molecular data for a comprehensive evolutionary analysis.

Main Methods:

• Developed a Bayesian approach modeling quantitative character evolution via Brownian diffusion, incorporating character correlation and within-population variation.
• Conducted simulations to evaluate the method's performance under varying correlation and population noise levels.
• Applied the method to craniofacial landmark data and molecular data from carnivoran mammals.

Main Results:

• Ignoring population variation and character correlation leads to biased estimates of divergence times and evolutionary rates.
• Divergence time estimates differ significantly based on the data type analyzed (morphological, molecular, or combined).
• Carnivoran morphological data exhibit high rate variation, with an independent-rates model fitting better than an autocorrelated-rates model.

Conclusions:

• The new Bayesian method provides a more robust estimation of divergence times by accounting for quantitative character complexities.
• Integrating quantitative morphological data with molecular data offers valuable insights into species evolution.
• The developed model, implemented in MCMCtree, enhances phylogenetic inference capabilities.