Related Experiment Videos
Detection of recombination in DNA multiple alignments with hidden Markov models
1Biomathematics and Statistics Scotland, SCRI, Invergowrie, Dundee, United Kingdom. dirk@bioss.sari.ac.uk
Summary
This study introduces a hidden Markov model (HMM) to detect recombination in DNA sequences, improving phylogenetic tree accuracy. The new method, using the expectation maximization (EM) algorithm, outperforms previous approaches in identifying evolutionary history changes.
Area of Science:
- Genomics
- Molecular Evolution
- Bioinformatics
Background:
- Phylogenetic tree estimation traditionally assumes a single evolutionary history for all DNA sites.
- Recombination in bacteria and viruses creates mosaic sequences, violating this assumption and leading to errors in phylogenetic analysis.
Purpose of the Study:
- To develop and validate a novel method for detecting recombination events in DNA multiple alignments.
- To improve the accuracy of phylogenetic tree estimation by accounting for recombination.
Main Methods:
- Utilized a hidden Markov model (HMM) incorporating emission probabilities based on phylogenetic tree topology and branch lengths.
- Employed the expectation maximization (EM) algorithm for maximum likelihood optimization of branch lengths and recombination probability.
- Compared the novel approach against an earlier heuristic method using a synthetic benchmark.
Main Results:
- The HMM-based EM algorithm significantly outperformed the heuristic method in detecting recombination events.
- The method successfully identified a likely recombination event in an argF gene alignment from Neisseria strains.
Conclusions:
- The developed HMM-EM algorithm provides a robust and accurate method for detecting recombination in DNA sequences.
- Accurate detection of recombination is crucial for reliable phylogenetic inference, especially in bacteria and viruses.