Related Experiment Videos
A Bayesian model for detecting past recombination events in DNA multiple alignments.
G McGuire1, F Wright, M J Prentice
1Biomathematics and Statistics Scotland, JCMB, Edinburgh.
Summary
Detecting recombination is crucial for accurate phylogenetic analysis. This study introduces a Bayesian Hidden Markov model to identify topological changes along sequence alignments, improving evolutionary relationship inference.
Area of Science:
- Evolutionary biology
- Bioinformatics
- Computational biology
Background:
- Phylogenetic tree estimation typically assumes a single evolutionary history for all sites in a sequence alignment.
- Recombination events create mosaic sequences, violating this assumption and potentially leading to inaccurate phylogenetic inferences.
- Accurate detection of past recombination is essential for reliable phylogenetic analysis.
Purpose of the Study:
- To develop and present a novel Bayesian model for detecting recombination events and estimating changes in phylogenetic topology along sequence alignments.
- To relax the assumption of a single topology across an entire alignment, accommodating mosaic sequence structures.
- To provide a robust method for identifying sites with differing evolutionary histories within a dataset.
Main Methods:
- A Bayesian model utilizing Hidden Markov models (HMMs) to infer varying phylogenetic topologies across a multiple sequence alignment.
- The HMM framework treats underlying phylogenetic topologies as hidden states, allowing for transitions between them.
- Maximum a posteriori (MAP) estimation is employed to determine the most probable changes in topology along the alignment.
Main Results:
- The developed Bayesian HMM framework successfully models and estimates changes in phylogenetic topology caused by recombination.
- The Maximum a posteriori (MAP) estimate provides a reliable method for identifying breakpoints and varying evolutionary histories.
- Performance assessment on simulated and real sequence data demonstrates the model's efficacy in detecting recombination.
Conclusions:
- The proposed Bayesian Hidden Markov model offers a significant advancement in phylogenetic analysis by accounting for recombination.
- This method improves the accuracy of evolutionary relationship inference by accommodating mosaic sequence structures.
- The model is a valuable tool for researchers needing to detect recombination and analyze complex evolutionary histories in sequence data.