GHOST: Recovering Historical Signal from Heterotachously Evolved Sequence Alignments.

Area of Science:

• Evolutionary biology
• Phylogenetics
• Bioinformatics

Background:

• Molecular sequence data evolution can be complex and misleading for phylogenetic inference.
• Traditional models often fail to account for heterotachous evolutionary processes.

Purpose of the Study:

• Introduce the General Heterogeneous evolution On a Single Topology (GHOST) model for phylogenetic inference.
• Evaluate the GHOST model's performance on simulated and empirical data.

Main Methods:

• Implemented the GHOST model within the IQ-TREE phylogenetic software.
• Utilized maximum-likelihood framework for phylogenetic analysis.
• Tested with simulated heterotachously evolved sequences and empirical phylogenomic data.

Main Results:

• GHOST model accurately recovers tree topology, branch lengths, and substitution model parameters.
• Phylogenetic analysis of turtles and archosaurs using GHOST concurs with accepted views.
• Identified convergent evolution signals in electric fish using the GHOST model.

Conclusions:

• The GHOST model provides accurate phylogenetic inference from heterotachous data.
• GHOST offers unique biological insights by minimizing model constraints compared to traditional methods.
• This model advances our understanding of evolutionary processes and relationships.