Incorporating indel information into phylogeny estimation for rapidly emerging pathogens

Area of Science:

• Computational Biology
• Evolutionary Biology
• Genomics

Background:

• Resolving phylogenies for rapidly evolving pathogens is challenging due to limited sequence divergence.
• Traditional methods often rely solely on substitution data, potentially missing crucial evolutionary signals.

Purpose of the Study:

• To enhance phylogenetic resolution for rapidly evolving pathogens by integrating insertion/deletion (indel) information.
• To address biases from single-alignment reliance by jointly estimating alignment and phylogeny.

Main Methods:

• Developed a novel Markov chain transition kernel for efficient joint estimation of alignment and phylogeny in a Bayesian framework.
• Extended indel models to incorporate biological realism, favoring indel placement on longer evolutionary branches.
• Utilized Markov chain Monte Carlo (MCMC) for inference, considering an ensemble of near-optimal alignments.

Main Results:

• Indel information significantly increased phylogenetic resolution in within-host viral sequence samples.
• Demonstrated the critical importance of accounting for alignment uncertainty in phylogenetic analyses.
• Showcased how codon-based substitution models can negatively impact alignment quality and phylogenetic inference.

Conclusions:

• Indel data offers a valuable tool for improving phylogenetic resolution of recently diverged pathogens.
• Accounting for alignment uncertainty is essential for accurate phylogenetic reconstruction when using indel information.