Inferring pattern and process: maximum-likelihood implementation of a nonhomogeneous model of DNA sequence evolution...

Area of Science:

• Evolutionary Biology
• Molecular Evolution
• Bioinformatics

Background:

• Traditional models often assume homogeneous base compositions across DNA sequences.
• Unequal base compositions (G+C content variation) among lineages can bias phylogenetic inference.
• Accurate reconstruction of evolutionary history requires models that account for such variations.

Purpose of the Study:

• To develop a nonhomogeneous, nonstationary stochastic model for DNA sequence evolution.
• To incorporate varying equilibrium G+C contents among lineages to handle unequal base compositions.
• To implement a maximum-likelihood framework for phylogenetic analyses with this new model.

Main Methods:

• Devised a novel stochastic model of DNA sequence evolution.
• Implemented a maximum-likelihood approach for phylogenetic analysis.
• Assessed model relevance and parameter accuracy using real and simulated DNA data sets.

Main Results:

• The model accurately estimates G+C content at ancestral nodes.
• It allows simultaneous inference of evolutionary process (substitution rates) and pattern (phylogenetic tree).
• Significant information about past evolutionary modes can be extracted from DNA sequences.

Conclusions:

• The developed model is effective for phylogenetic reconstruction with varying base compositions.
• It offers a valuable tool for molecular evolution studies.
• The method enhances the accuracy of inferring evolutionary relationships and ancestral states.