Molecular timetrees using relaxed clocks and uncertain phylogenies.

Area of Science:

• Evolutionary Biology
• Computational Biology
• Bioinformatics

Background:

• Molecular systematics commonly infers phylogeny then estimates divergence times using relaxed clock methods.
• This sequential approach overlooks phylogenetic uncertainty's impact on divergence time estimates and credibility intervals.
• Joint inference of phylogeny and divergence times offers an alternative to incorporate phylogenetic errors into molecular dating.

Purpose of the Study:

• To compare the performance of sequential versus joint analyses in reconstructing evolutionary timetrees.
• To evaluate the impact of phylogenetic uncertainty on divergence time estimates.
• To develop a computationally efficient method for joint phylogenetic and divergence time inference.

Main Methods:

• Computer-simulated and empirical datasets were used for comparative analyses.
• Bayesian methods for joint inference were assessed for computational feasibility.
• A novel joint inference approach combining bag of little bootstraps, maximum likelihood, and RelTime was developed.

Main Results:

• Sequential and joint analyses yielded similar divergence times and phylogenetic relationships in most cases.
• Joint inference demonstrated superior performance when phylogenies were not well resolved.
• The new joint inference method significantly reduced computational burden compared to existing Bayesian approaches.

Conclusions:

• Joint inference is preferable for molecular dating when phylogenetic uncertainty is high or phylogenies are poorly resolved.
• The developed method provides a computationally efficient alternative for joint phylogenetic and divergence time inference.
• This approach effectively incorporates phylogenetic uncertainty into molecular dating, yielding robust results.