Dating phylogenies with hybrid local molecular clocks

Area of Science:

• Computational Biology
• Evolutionary Biology
• Phylogenetics

Background:

• Estimating species divergence times from molecular data requires assumptions about evolutionary rates.
• Existing methods like molecular clock models, penalized likelihood, and Bayesian approaches often underestimate rate variation, impacting date accuracy.

Purpose of the Study:

• To develop and evaluate novel methods for estimating molecular divergence times.
• To address limitations in current molecular clock models by incorporating phylogenetic uncertainty and local clock approaches.

Main Methods:

• A modified rate-smoothing algorithm automatically places local molecular clocks on phylogenies.
• Hybrid approaches integrating microarray data analysis techniques are employed.
• Phylogenetic uncertainty is integrated within local clock models.

Main Results:

• The new maximum likelihood hybrid methods demonstrate improved performance compared to penalized likelihood.
• These methods perform comparably to uncorrelated Bayesian models in estimating divergence times.
• A tendency to underestimate the extent of evolutionary rate change persists.

Conclusions:

• The developed hybrid methods offer a more accurate approach to molecular clock dating.
• Despite improvements, accurately estimating divergence times using local molecular clocks remains challenging due to rate variation.
• This study underscores the inherent difficulties in molecular dating and suggests avenues for future refinement.