Expansion of biological pathways based on evolutionary inference.

Area of Science:

• Genomics
• Bioinformatics
• Evolutionary Biology

Background:

• Predicting gene function often relies on shared evolutionary history across species.
• Pathways composed of distinct evolutionary modules present challenges for traditional functional prediction methods.

Purpose of the Study:

• To introduce a novel computational algorithm, CLIME (Clustering by Inferred Models of Evolution), for identifying evolutionary modules within gene sets.
• To develop a method that simultaneously learns the number of modules and their evolutionary histories.
• To expand identified modules by discovering new components that fit the inferred evolutionary models.

Main Methods:

• CLIME algorithm takes a eukaryotic species tree, homology matrix, and gene set as input.
• It partitions the gene set into disjoint evolutionary modules.
• It infers a tree-based evolutionary history for each module and scans genomes for new components.

Main Results:

• CLIME identified unanticipated evolutionary modularity in approximately 1,000 human pathways.
• The algorithm revealed coevolving components within these pathways across yeast, red algae, and malaria proteomes.
• Demonstrated the ability to discover novel pathway members based on inferred evolutionary models.

Conclusions:

• CLIME effectively reveals hidden evolutionary structures within biological pathways.
• The algorithm enhances the prediction of gene function by accounting for evolutionary modularity.
• CLIME's utility is expected to grow with increasing availability of eukaryotic genome data.