Prediction of tissue-specific cis-regulatory modules using Bayesian networks and regression trees
2007

Predicting Tissue-Specific Gene Regulation

Sample size: 6278 publication Evidence: high

Author Information

Author(s): Chen Xiaoyu, Blanchette Mathieu

Primary Institution: McGill Centre for Bioinformatics

Hypothesis

Can a Bayesian network approach accurately identify cis-regulatory modules that regulate tissue-specific gene expression?

Conclusion

The developed method accurately identifies known human liver and erythroid-specific modules and predicts important transcription factor combinations associated with specific expression in various tissues.

Supporting Evidence

  • The method accurately discriminates between known liver and erythroid-specific modules.
  • Important combinations of transcription factors were predicted to be linked to tissue-specific expression.
  • The approach integrates multiple data types to enhance prediction accuracy.

Takeaway

The study created a computer program that helps find parts of DNA that control how genes work in different tissues, like the liver or blood.

Methodology

A Bayesian network was developed to integrate transcription factor binding site predictions and tissue-specific expression data to identify cis-regulatory modules.

Potential Biases

The reliance on expression data may introduce biases if the data is inaccurate or incomplete.

Limitations

The method may misidentify false-positive regulatory modules and does not account for unobserved factors affecting regulatory activity.

Digital Object Identifier (DOI)

10.1186/1471-2105-8-S10-S2

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication