Evaluating Models for Phylogenetic Inference
Author Information
Author(s): Zhou Yan, Rodrigue Nicolas, Lartillot Nicolas, Philippe Hervé
Primary Institution: Université de Montréal
Hypothesis
How do covarion and mixture of branch lengths (MBL) models compare in handling heterotachy in phylogenetic inference?
Conclusion
The covarion model is more efficient at handling heterotachy than the MBL model.
Supporting Evidence
- The study found that the covarion model is always favored over the optimal MBL model.
- Cross-validation was shown to be more reliable than AIC and BIC for model comparison.
- Simulations indicated that the MBL model requires a significant increase in parameters compared to the covarion model.
Takeaway
This study looks at different ways to understand how species evolve over time. It finds that one method works better than another for certain types of data.
Methodology
The study used simulations and analyzed three large datasets to compare the performance of covarion and MBL modeling strategies.
Limitations
The MBL model is computationally expensive and may not perform well with weak heterotachous signals.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website