The effect of heterotachy in multigene analysis using the neighbor joining method

Area of Science:

• Phylogenetics and evolutionary biology.
• Bioinformatics and computational biology.

Background:

• Multigene sequence alignments are crucial for inferring species phylogenies.
• Phylogenetic accuracy relies on the assumption of homotachy, where evolutionary rates are constant within lineages for each gene.
• Heterotachy, or within-site rate variation, violates this assumption and can lead to inaccurate phylogenetic reconstructions.

Purpose of the Study:

• To investigate the impact of heterotachy on species phylogenies reconstructed using concatenated gene data.
• To evaluate the performance of the neighbor joining (NJ) method under heterotachy.
• To develop and assess methods that mitigate the effects of heterotachy in multigene phylogenetic analyses.

Main Methods:

• Theoretical examination of a four-taxon case using the neighbor joining (NJ) method.
• Simulation of concatenated gene datasets exhibiting heterotachy.
• Application and evaluation of average and weighted-average distance approaches that maintain gene boundaries.

Main Results:

• The neighbor joining (NJ) method incorrectly reconstructs the species phylogeny when concatenated genes display heterotachy.
• Heterotachy significantly distorts phylogenetic relationships inferred from standard concatenation methods.
• Average and weighted-average distance methods effectively overcome the detrimental effects of heterotachy.

Conclusions:

• Violation of the homotachy assumption via heterotachy poses a significant challenge for accurate phylogenetic inference.
• Standard concatenation and NJ methods are susceptible to errors caused by heterotachy.
• Preserving gene boundaries using average or weighted-average distance approaches offers a robust solution for heterotachy in multigene phylogenetics.