Related Experiment Videos
A statistical model for locating regulatory regions in genomic DNA
E M Crowley1, K Roeder, M Bina
1Department of Statistics, Carnegie Mellon University, Pittsburgh, PA 15213-3890, USA.
Journal of Molecular Biology
|April 25, 1997
Summary
This study introduces a Bayesian model to identify gene regulatory regions in DNA by analyzing protein-binding elements. The model effectively pinpoints these crucial regulatory DNA sequences.
Area of Science:
- Genomics
- Bioinformatics
- Computational Biology
Background:
- Chromosomal DNA contains regulatory sequences that control gene expression.
- These regulatory signals are often found in clusters within specific genomic regions.
Purpose of the Study:
- To develop a novel Bayesian model for identifying regulatory regions in genomic DNA.
- To treat the localization of regulatory regions as a change-point problem within DNA sequences.
Main Methods:
- Utilizing a hidden Markov chain model.
- Analyzing nucleotide positions of protein-binding elements identified via a reference catalog of transcription factor interactions.
- Employing a statistical model to automatically select predictive protein-binding elements for regulatory regions.
Main Results:
- The Bayesian model successfully identifies regulatory regions in both viral and human genomic DNA.
- Demonstrated efficacy on the beta globin locus on human chromosome 11.
- The model automatically selects relevant protein-binding elements for accurate prediction.
Conclusions:
- The developed Bayesian change-point model is effective for locating gene regulatory regions.
- This approach enhances the understanding of gene expression control mechanisms.
- The model provides a robust tool for genomic sequence analysis.