Searching ChIP-seq genomic islands for combinatorial regulatory codes in mouse embryonic stem cells

Area of Science:

• Genomics
• Stem Cell Biology
• Transcriptional Regulation

Background:

• Mouse embryonic stem (ES) cells possess complex transcriptional regulatory circuits.
• Key transcription factors (TFs) binding sites have been mapped using ChIP-seq.
• Gene expression profiles in ES cells and differentiated states are known and correlate with TF binding.

Purpose of the Study:

• To investigate the role of potential cofactors in gene regulation within mouse ES cells.
• To identify combinatorial genomic features associated with gene expression patterns.
• To determine if cofactors collaborate with known TFs.

Main Methods:

• Analysis of ES cell gene expression profiles and TF binding data.
• Identification of cofactor binding sites near known TF binding locations.
• Statistical analysis to identify co-binding features associated with differential gene expression (activated vs. repressed).
• Gene classification using identified features and comparison with TF-only classification.

Main Results:

• A list of co-binding features significantly associated with gene expression patterns was identified at a 10% false discovery rate.
• Gene classification accuracy improved by up to 20% when incorporating identified features compared to using ChIP-seq TFs alone.
• Over one-third of the proposed cofactor roles are supported by existing literature.
• Predicted target genes for most candidate cofactors showed expected expression changes in an independent dataset, validating the findings.

Conclusions:

• Combinatorial genomic features associated with gene expression in ES cells were revealed.
• These features suggest the involvement of potential cofactors collaborating with ChIP-seq TFs in gene regulation.
• The study provides a foundation for further investigation into cofactor-mediated transcriptional control in stem cells.