Broccoli: Combining Phylogenetic and Network Analyses for Orthology Assignment.

Area of Science:

• Bioinformatics
• Comparative Genomics
• Computational Biology

Background:

• Orthology assignment is crucial for comparative genomics.
• Existing gene clustering pipelines rely on protein distances, which can be prone to artifacts.
• There is a need for more precise and reliable orthology assignment methods.

Purpose of the Study:

• Introduce Broccoli, a user-friendly pipeline for accurate orthology assignment.
• Develop a phylogeny-based approach to overcome limitations of distance-based methods.
• Enable detection and assignment of chimeric proteins.

Main Methods:

• Broccoli utilizes ultrafast phylogenetic analyses to build a network of orthologous relationships.
• A parameter-free machine learning algorithm identifies orthologous groups from the network.
• The pipeline is designed to detect chimeric proteins resulting from gene-fusion events.

Main Results:

• Broccoli demonstrates superior performance compared to current orthology pipelines on benchmark datasets.
• The pipeline achieves high precision in inferring orthologous groups and pairs.
• Broccoli exhibits scalability with runtimes comparable to distance-based pipelines.

Conclusions:

• Broccoli offers a precise, efficient, and user-friendly solution for orthology assignment.
• Its performance makes it a suitable choice for large-scale comparative genomic studies.
• The pipeline effectively handles complex cases like chimeric proteins.