Bayesian Analysis for Rheumatoid Arthritis Genetics
Author Information
Author(s): Fredrick R Schumacher, Peter Kraft
Primary Institution: Harvard School of Public Health
Hypothesis
The Bayesian analysis might be better able to detect true associations while minimizing false positives.
Conclusion
Bayesian latent class analyses provided no obvious improvement over maximum-likelihood estimation.
Supporting Evidence
- The Bayesian analysis can borrow strength from multiple loci to estimate association parameters.
- The posterior means for the log allelic odds ratios were less variable than the maximum likelihood estimates.
- The posterior probabilities were not as good as the simple p-values in distinguishing a signal from a non-signal.
Takeaway
Researchers used a new method to find genetic links to rheumatoid arthritis but didn't find it worked better than older methods.
Methodology
The study used a Bayesian latent class analysis and maximum-likelihood estimation on simulated data from rheumatoid arthritis cases and controls.
Potential Biases
The prior probabilities may be overwhelmed by the large amount of data.
Limitations
The results may not be generalizable due to the large effect simulated in the human leukocyte antigen-DR locus.
Participant Demographics
1500 rheumatoid arthritis cases and 2000 controls.
Want to read the original?
Access the complete publication on the publisher's website