Building an adaptive trait simulator package to infer parametric diffusion model along phylogenetic tree.

Area of Science:

• Evolutionary Biology
• Phylogenetics
• Computational Biology

Background:

• Modeling trait evolution along phylogenetic trees is crucial for understanding species diversification.
• Continuous-time stochastic processes are widely used but require robust inference methods.
• Existing methods may lack flexibility or require complex likelihood calculations.

Purpose of the Study:

• To develop an adaptive trait simulator package for inferring trait evolution.
• To incorporate the Cox-Ingersol-Ross process for realistic rate evolution modeling.
• To provide a feasible statistical inference framework using approximate Bayesian computation.

Main Methods:

• Development of a trait simulator package for phylogenetic trees.
• Implementation of continuous-time stochastic processes, including the Cox-Ingersol-Ross process.
• Integration of approximate Bayesian computation (ABC) for model inference without likelihood.

Main Results:

• The package enables the simulation of trait evolution under various stochastic processes.
• The Cox-Ingersol-Ross process inclusion prevents negative evolutionary rates, suitable for adaptive evolution.
• The ABC procedure offers a viable alternative for statistical inference in phylogenetics.

Conclusions:

• The adaptive trait simulator package provides a flexible tool for evolutionary data analysis.
• This approach facilitates the study of adaptive trait evolution in dynamic environments.
• The method is broadly applicable to existing trait evolution models, offering users a valuable alternative.