A Fundamental Problem with Amino-Acid-Sequence Characters for Phylogenetic Analyses.

Area of Science:

• Evolutionary biology
• Molecular evolution
• Bioinformatics

Background:

• Phylogenetic analyses commonly utilize nucleotide or amino acid sequences.
• Amino acid sequences are often preferred for their ability to correct for saturation (parallelism).

Purpose of the Study:

• To investigate the impact of using amino acid sequence characters versus nucleotide sequence characters in phylogenetic analyses.
• To identify potential issues arising from amino acid coding, such as convergence and loss of informative variation.

Main Methods:

• Comparative analysis of phylogenetic character coding using nucleotide and amino acid sequences.
• Evaluation of homoplasy and phylogenetic signal using consistency index.
• Examination of tree-construction method outcomes under different coding schemes.

Main Results:

• Amino acid characters, while correcting for saturation, are prone to convergence, potentially masking true evolutionary relationships.
• Nucleotide sequence characters can retain phylogenetically informative variation that may be lost or obscured in amino acid coding.
• Differences in coding can lead to conflicting phylogenetic groupings, affecting the reliability of evolutionary inferences.

Conclusions:

• Coding amino acid sequences instead of nucleotide sequences can introduce convergence problems in phylogenetic analyses.
• The choice of character coding significantly impacts the accuracy and reliability of phylogenetic tree reconstruction.
• Careful consideration of character delimitation is crucial for robust phylogenetic inference, as demonstrated by the Mhc locus example.