Evaluating the Performance of Widely Used Phylogenetic Models for Gene Expression Evolution

Area of Science:

• Evolutionary biology
• Genomics
• Bioinformatics

Background:

• Phylogenetic comparative methods (PCMs) are crucial for studying gene expression evolution.
• The suitability of standard phylogenetic models for gene expression data remains unclear.
• Model fit directly impacts the reliability of evolutionary conclusions.

Approach:

• Fitted multiple phylogenetic models of trait evolution to eight comparative gene expression datasets.
• Evaluated model performance using a previously established statistical approach.
• Assessed how well the best-fit model described the expression data in an absolute sense.

Key Points:

• Ornstein-Uhlenbeck models, reflecting constrained expression optima, were favored for 66% of gene-tissue combinations.
• The best-fit model performed adequately for 61% of gene-tissue combinations.
• Poor model performance often stemmed from unaddressed evolutionary rate heterogeneity.

Conclusions:

• Model fit assessment should be standard practice in phylogenetic comparative expression studies.
• Ensuring model adequacy enhances the reliability of evolutionary inferences.
• Identifying model limitations can drive the development of more sophisticated evolutionary models.