Phylogenetic Comparative Methods on Phylogenetic Networks with Reticulations

Area of Science:

• Evolutionary Biology
• Phylogenetics
• Quantitative Genetics

Background:

• Phylogenetic comparative methods (PCMs) traditionally model trait evolution on trees, assuming Brownian Motion (BM).
• Reticulate evolution (hybridization, gene flow) is common but not captured by tree-based models.
• Phylogenetic networks are emerging to represent reticulate evolution, necessitating new trait evolution models.

Purpose of the Study:

• To develop novel PCMs applicable to phylogenetic networks.
• To extend existing PCM tools (regression, ancestral reconstruction, Pagel's lambda test) to networks.
• To model and test for transgressive evolution following reticulation events.

Main Methods:

• Developed an efficient recursive algorithm for computing trait variance on networks.
• Extended standard PCMs to phylogenetic networks, including regression and ancestral reconstruction.
• Designed statistical tests for ancestral transgressive evolution using a regression framework.

Main Results:

• The new methods efficiently compute trait variance and enable standard PCM analyses on networks.
• Statistical tests for transgressive evolution were developed and their power analyzed.
• Recent reticulation events were shown to have the strongest impact on trait distribution.

Conclusions:

• The developed framework provides a powerful new approach for studying trait evolution in the presence of reticulation.
• The methods are applicable to diverse evolutionary scenarios, including transgressive evolution.
• The methods were successfully applied to Xiphophorus fishes and are available in the PhyloNetworks package.