Transcriptional regulatory network discovery via multiple method integration: application to e. coli K12

Area of Science:

• Computational Biology
• Systems Biology
• Bioinformatics

Background:

• Transcriptional regulatory network (TRN) discovery is often incomplete or inaccurate when using single data sources.
• Existing methods struggle with integrating diverse biological data effectively.

Purpose of the Study:

• To develop a novel methodology, TRND, for accurate TRN discovery by integrating multiple data types.
• To improve the comprehensiveness and accuracy of discovered TRNs.

Main Methods:

• TRND integrates preliminary TRNs, microarray data, gene ontology, and phylogenic similarity.
• A Bayesian framework facilitates multi-method integration using individual scoring measures and a training set.
• Gene-gene scores are transformed to gene-transcription factor (TF) scores using a preliminary TRN.

Main Results:

• TRND accurately discovers TRNs, demonstrated in E. coli K12.
• The integrated approach yields TRNs in better agreement with microarray data.
• TRND discovers double the number of gene/TF interactions compared to existing networks.

Conclusions:

• TRND offers a robust and extensible framework for TRN discovery.
• Transforming gene-gene scores to gene-TF scores enhances information extraction.
• This multi-method integration approach significantly improves TRN discovery accuracy and completeness.