Gene tree rooting methods give distributions that mimic the coalescent process.

Area of Science:

• Phylogenetics
• Evolutionary Biology
• Computational Biology

Background:

• Multi-locus phylogenetic inference commonly uses coalescent models to account for incomplete lineage sorting.
• Incomplete lineage sorting can cause gene trees to incongruent with the species tree.
• The "goodness-of-fit" of empirical data to coalescent models is an area of ongoing research.

Purpose of the Study:

• To investigate whether gene tree rooting methods can explain data variation attributed to the coalescent process.
• To assess the impact of different rooting strategies (molecular clock vs. outgroup rooting) on gene tree distributions.
• To evaluate the performance of coalescent-based species tree inference methods under these conditions.

Main Methods:

• Utilized simulated multi-locus sequence data for a four-taxon phylogenetic scenario.
• Estimated gene trees and applied both molecular clock and outgroup rooting methods.
• Compared observed gene tree distributions with predictions from coalescent models.

Main Results:

• For four taxa, gene tree distributions resulting from molecular clock or outgroup rooting closely matched coalescent model predictions.
• The choice of species tree branch lengths significantly influenced the match between observed and predicted gene tree distributions.
• Coalescent-based species tree inference methods showed varied performance when data "fit" the model due to rooting artifacts.

Conclusions:

• Gene tree rooting can be a significant factor in observed gene tree distributions, potentially mimicking coalescent effects.
• Careful consideration of rooting methods is crucial when evaluating the fit of data to coalescent models.
• The performance of species tree inference methods may be compromised if gene tree incongruence is an artifact of rooting rather than true biological processes.