Assessing an unknown evolutionary process: effect of increasing site-specific knowledge through taxon addition

Area of Science:

• Molecular Evolution
• Bioinformatics
• Computational Biology

Background:

• Accurate assessment of evolutionary processes is vital for understanding protein evolution and molecular analyses.
• Taxon sampling and sequence length are key factors influencing phylogenetic inference accuracy.

Purpose of the Study:

• To investigate how taxon sampling impacts parameter estimation and topological inference in molecular evolution simulations.
• To differentiate the effects of increased knowledge about internal nodes versus site-specific rates on phylogenetic accuracy.

Main Methods:

• Utilized simulation studies to model evolutionary processes.
• Employed maximum-likelihood analysis to assess phylogenetic models and parameters.
• Developed and applied the "doppelgänger trees" method to dissect sources of topological inference improvement.

Main Results:

• Increased taxon sampling substantially enhances support for evolutionary models and parameter accuracy, outperforming increased sequence length.
• Distinguished contributions of internal node knowledge and site-specific rate parameter accuracy to topological inference.
• Found that high model support alone does not guarantee improved topological inference; accurate site-specific rate assessment is crucial.

Conclusions:

• Extensive taxon sampling is more effective than increased sequence length for improving phylogenetic analyses, particularly when evolutionary rates vary across sites.
• Accurate estimation of site-specific evolutionary processes is essential for robust topological inference.
• Findings suggest prioritizing broader taxonomic sampling in molecular evolution studies, especially for datasets with moderate sequence lengths.