Choosing the Right Protein Models for Phylogenetic Analysis
Author Information
Author(s): Keane Thomas M, Creevey Christopher J, Pentony Melissa M, Naughton Thomas J, Mclnerney James O
Primary Institution: National University of Ireland, Maynooth
Hypothesis
Is it appropriate to choose protein models based on their source or method of construction?
Conclusion
Choosing protein models based on their source or method of construction may not be appropriate.
Supporting Evidence
- The study demonstrated that different protein models can produce significantly different phylogenetic trees.
- Statistical methods of model selection were shown to be robust and useful for protein model selection.
- No single amino acid matrix was found to be optimal for all datasets analyzed.
Takeaway
This study shows that picking protein models randomly can lead to very different results, so we should be careful about how we choose them.
Methodology
The study used simulations and real datasets to investigate the effects of different protein models on phylogenetic analysis.
Potential Biases
Ad hoc model selection may lead to misleading phylogenetic trees.
Limitations
The study does not address how well the selected models represent the actual evolutionary processes.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website