BranchClust: A Phylogenetic Algorithm for Selecting Gene Families
Author Information
Author(s): Maria S. Poptsova, J. Peter Gogarten
Primary Institution: University of Connecticut
Hypothesis
We aimed to develop a method of assembling orthologous gene families that would have no restriction on the number of taxa, doesn't require a known species tree and would be able to distinguish between paralogs and orthologs by analyzing their position in a phylogenetic tree.
Conclusion
BranchClust outperforms the Reciprocal Best Blast hit method in selecting more sets of putatively orthologous genes.
Supporting Evidence
- BranchClust can select both complete and incomplete clusters of orthologous genes.
- The algorithm is implemented in Perl and is freely available.
- BranchClust was tested on various sets of genomes, demonstrating its effectiveness.
Takeaway
BranchClust is a computer program that helps scientists find and group similar genes from different species, making it easier to study how these genes evolved.
Methodology
The BranchClust algorithm uses a phylogenetic approach to identify and group orthologous genes from multiple species without needing a known species tree.
Limitations
The algorithm's performance may be affected by the choice of starting genome and the presence of paralogs.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website