PSA: an effective method for predicting horizontal gene transfers through parsimonious phylogenetic networks

Area of Science:

• Evolutionary biology
• Bioinformatics
• Computational biology

Background:

• Horizontal gene transfer (HGT) plays a significant role in species evolution.
• Phylogenetic networks are used to model HGTs, often relying on parsimony values.
• Current parsimony definitions assume independent gene site evolution, which may not reflect biological reality.

Purpose of the Study:

• To define a novel parsimony value for phylogenetic networks that considers a gene evolving as a whole.
• To develop an algorithm for constructing phylogenetic networks based on this new parsimony definition.
• To improve the accuracy of predicting HGT events in evolutionary studies.

Main Methods:

• Introduction of a novel parsimony definition (p definition) for phylogenetic networks.
• Development of the Phylogeny with Simulated Annealing (PSA) algorithm.
• Utilizing Simulated Annealing optimization technique to search for optimal phylogenetic networks.

Main Results:

• The PSA algorithm successfully constructed parsimonious phylogenetic networks.
• Networks generated by PSA better represent evolutionary relationships involving HGTs.
• HGT predictions using PSA demonstrated higher accuracy compared to existing methods.

Conclusions:

• The novel p definition and PSA algorithm offer a more biologically realistic approach to phylogenetic network construction.
• PSA enhances the accuracy of HGT prediction, crucial for understanding evolutionary processes.
• The PSA algorithm is publicly available for further research and application.