Tracer

Tracer visualizes and diagnoses Markov chain Monte Carlo (MCMC) output from BEAST to assess convergence, mixing, and summarize posterior estimates for molecular sequence-based evolutionary analyses.

Key Features:

• MCMC Output Visualization: Produces graphical representations of MCMC output including trace plots and density plots for interpreting parameter trajectories and posterior distributions.
• Diagnostics for Convergence and Mixing: Computes diagnostic measures and visual diagnostics to evaluate chain convergence and mixing and to determine whether chains have adequately sampled the posterior.
• Compatibility with BEAST 1.7: Reads output files generated by BEAST version 1.7 for integration with Bayesian phylogenetic analyses.
• Support for Advanced Models via BEAUti: Interfaces with BEAUti to access advanced models for molecular sequence evolution and phenotypic trait evolution specified in BEAST input files.

Scientific Applications:

• Bayesian Phylogenetic Inference: Supports assessment and interpretation of parameter estimates and posterior distributions from Bayesian phylogenetic analyses.
• Divergence Time Dating: Aids evaluation of posterior estimates used to infer divergence times among taxa.
• Coalescent Analysis: Supports diagnostics and summarization of MCMC output from coalescent-based population genetic models.
• Phylogeography: Facilitates inspection of posterior estimates relevant to phylogeographic inference of geographic and demographic history.

Methodology:

Reads BEAST 1.7 output files; generates trace plots and density plots; computes diagnostic measures for convergence and mixing; and interfaces with BEAUti to obtain model specifications for visualization and diagnostics.