Rate-time scaling in phenotypic evolution: Limitations of current models in capturing temporal dynamics

Area of Science:

• Evolutionary biology
• Phylogenetics
• Quantitative genetics

Background:

• Evolutionary rates are crucial for understanding diversification but often correlate negatively with time.
• This rate-time correlation complicates comparisons across lineages with different diversification histories.
• The underlying causes of this rate-time scaling remain debated in evolutionary studies.

Purpose of the Study:

• To investigate the theoretical possibility of estimating time-independent evolutionary rates.
• To assess the impact of common analytical issues on the rate-time correlation.
• To determine if empirical data supports time-independent rate estimation.

Main Methods:

• Computer simulations using the unbiased random walk (Brownian motion) model.
• Analysis of 643 empirical time series of trait evolution.
• Evaluation of factors like model misspecification, sampling error, and model identifiability.

Main Results:

• Simulations show that Brownian motion models can theoretically yield time-independent rate estimates.
• Empirical analyses indicate that model misspecification, sampling error, and identifiability issues do not significantly reduce the negative rate-time scaling.
• Commonly used models may not accurately capture empirical trait evolution.

Conclusions:

• The negative correlation between evolutionary rates and time likely requires evolutionary explanations beyond analytical artifacts.
• Current models in phylogenetic comparative studies and phenotypic time series analyses may be inadequate for empirical data.
• Accurate comparison of evolutionary rates across lineages with varying timescales remains a significant challenge.