Identification of functional modules based on transcriptional regulation structure

Area of Science:

• Genomics
• Systems Biology
• Bioinformatics

Background:

• Identifying gene functional modules is crucial for understanding gene functions on a large scale.
• Current clustering algorithms primarily use gene co-expression, grouping genes with similar expression patterns.

Purpose of the Study:

• To propose and develop a novel gene clustering approach based on co-regulation rather than co-expression.
• To infer regulatory relationships and cluster genes based on this inferred structure.
• To validate the effectiveness of the proposed clustering method using Gene Ontology (GO) enrichment.

Main Methods:

• An inference algorithm was developed to detect co-regulated gene groups from gene expression data.
• A method was introduced to cluster genes utilizing the inferred regulatory structure.
• Clustering validation was performed using a GO enrichment score for the identified gene groups.

Main Results:

• The proposed co-regulation-based clustering method was successfully implemented.
• The method effectively infers regulatory relationships and clusters genes accordingly.
• Validation using GO enrichment demonstrated the biological relevance of the clustered gene groups.

Conclusions:

• The developed methods were evaluated on Saccharomyces cerevisiae stress response data.
• Co-regulation-based clustering yielded superior results compared to direct co-expression clustering.
• This approach offers a more effective strategy for identifying functional gene modules and understanding gene regulation.