Maximum-Likelihood Tree Estimation Using Codon Substitution Models with Multiple Partitions.

Area of Science:

• Molecular Evolution
• Phylogenetics
• Bioinformatics

Background:

• Protein sequences often contain domains with varying evolutionary rates and selective pressures.
• Traditional phylogenetic models struggle to account for these domain-specific evolutionary differences.
• Codon bias can also vary across different regions of a protein sequence.

Purpose of the Study:

• To develop and test a multipartition approach for maximum-likelihood phylogeny inference.
• To allow the use of different codon models for predefined partitions within a dataset.
• To investigate the evolutionary origins of leucine-rich repeat regions in Ralstonia solanacearum type III effector proteins.

Main Methods:

• Implemented a multipartition approach for phylogenetic analysis.
• Applied different codon models to distinct partitions of protein sequence data.
• Tested the method with simulated data and a real-world case study of bacterial pathogens.

Main Results:

• A simple two-partition model significantly improved phylogenetic resolution compared to a one-partition model.
• The multipartition approach provided stronger evidence for lateral gene transfer events.
• Successfully resolved phylogenetic relationships in the study of Ralstonia solanacearum.

Conclusions:

• The multipartition approach offers a more robust method for phylogenetic inference with heterogeneous sequence data.
• This approach enhances the study of complex evolutionary processes, including lateral gene transfer.
• Provides new insights into the evolution of type III effector proteins in plant-pathogenic bacteria.