Bayesian Optimal Discovery Procedure for Simultaneous Significance Testing
Author Information
Author(s): Cao Jing, Xie Xian-Jin, Zhang Song, Whitehurst Angelique, White Michael A
Primary Institution: Southern Methodist University
Hypothesis
Can a Bayesian hierarchical model improve the performance of the optimal discovery procedure in high throughput screening?
Conclusion
The Bayesian ODP outperforms traditional methods, especially when there are few replicates per test.
Supporting Evidence
- The Bayesian ODP was shown to have optimal performance in multiple significance tests.
- Simulation studies indicated that the Bayesian ODP significantly outperformed the original ODP.
- The method effectively borrows strength across genes to improve parameter estimation.
Takeaway
This study shows that using a special math model can help scientists find important genes more accurately, even when they don't have many tests to compare.
Methodology
A Bayesian hierarchical model was developed and compared with several competing test statistics through simulation studies and real datasets.
Limitations
The performance of the Bayesian ODP may vary depending on the number of replicates and the specific characteristics of the datasets used.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website