PrioriTree: a utility for improving phylodynamic analyses in BEAST

Area of Science:

• Epidemiology
• Computational Biology
• Bioinformatics

Background:

• Phylodynamic methods are crucial for understanding disease outbreak origins and spread.
• Discrete-geographic phylodynamic models require careful parameter inference, making them sensitive to prior assumptions.

Purpose of the Study:

• To introduce PrioriTree, an interactive R package designed to identify and manage prior sensitivity in discrete-geographic phylodynamic analyses.
• To facilitate robust Bayesian inference and model selection for disease evolution studies.

Main Methods:

• PrioriTree generates input files for BEAST analyses.
• It summarizes output from BEAST, enabling data-cloning and model fit assessments.
• The package supports the evaluation of candidate discrete-geographic models against empirical data.

Main Results:

• PrioriTree provides a user-friendly interface for complex phylodynamic analyses.
• It enhances the reliability of inferences regarding geographic history and demographic factors in disease outbreaks.
• The tool aids in selecting appropriate prior models for phylodynamic inference.

Conclusions:

• PrioriTree empowers researchers to perform more rigorous phylodynamic analyses by addressing prior sensitivity.
• This utility improves the accuracy and interpretability of geographic and demographic history reconstructions for infectious diseases.