Stochastic Character Mapping of Continuous Traits on Phylogenies

Area of Science:

• Evolutionary Biology
• Computational Biology
• Phylogenetics

Background:

• Fossil records incompletely represent evolutionary history.
• Ancestral state reconstruction infers unobserved traits from relatives.
• Stochastic character mapping is effective for discrete traits.

Purpose of the Study:

• Generalize stochastic character mapping for continuous variables.
• Develop a method for inferring relationships between trait evolution rates and continuous factors.
• Introduce the `contsimmap` R package for continuous stochastic character mapping.

Main Methods:

• Developed an efficient algorithm for sampling evolutionary histories under Brownian Motion models.
• Incorporated multiple correlated phenotypes, measurement error, and lineage-specific variation.
• Created a novel pipeline for analyzing continuous trait evolution rates.

Main Results:

• Demonstrated the `contsimmap` package's utility on simulated and real data.
• Showed smaller eucalypt trees exhibit higher rates of flower and leaf trait evolution.
• Aligned findings with life history theory and empirical patterns in other species.

Conclusions:

• Continuous stochastic character mapping is a valuable tool for macroevolutionary data analysis.
• Enables rigorous investigation of complex evolutionary dynamics with continuous traits.
• Facilitates studying continuous variables influencing evolutionary processes.