Phylogenetic Comparative Methods for Evaluating the Evolutionary History of Function-Valued Traits.

Area of Science:

• Evolutionary Biology
• Phylogenetics
• Quantitative Genetics

Background:

• Phylogenetic comparative methods typically assume fixed trait values within species.
• Existing methods struggle to account for trait variation driven by environmental or temporal gradients (e.g., reaction norms, ontogenetic trajectories).
• Function-valued traits are described by mathematical functions linking predictor variables to the trait of interest.

Purpose of the Study:

• To introduce a novel method for extending ancestral state reconstruction to incorporate function-valued traits within a phylogenetic generalized least squares (PGLS) framework.
• To develop extensions for testing phylogenetic signal, performing phylogenetic analysis of variance (ANOVA), and assessing correlated trait evolution using multivariate PGLS.
• To compare the statistical power of function-valued comparative methods against univariate approaches via simulations.

Main Methods:

• Development of a function-valued trait extension for PGLS.
• Application of the method to ancestral state reconstruction, phylogenetic signal testing, phylogenetic ANOVA, and correlated trait evolution analyses.
• Data simulations to compare statistical power between function-valued and univariate comparative methods.

Main Results:

• A new PGLS-based framework is presented for analyzing function-valued traits in a phylogenetic context.
• The method allows for more nuanced investigations into trait evolution, accommodating complex trait dynamics.
• Simulations provide insights into the statistical power and assumptions of these novel comparative methods.

Conclusions:

• The proposed method significantly advances phylogenetic comparative analyses by enabling the study of function-valued traits.
• This framework offers a powerful tool for understanding trait evolution influenced by environmental or temporal factors.
• Further research should explore the assumptions and challenges associated with applying these advanced methods.