Evaluating False Discovery Rate in Genetic Studies
Author Information
Author(s): Meredith E Tabangin, Jessica G Woo, Chunyan Liu, Todd G Nick, Lisa J Martin
Primary Institution: Cincinnati Children's Hospital Medical Center
Hypothesis
How does the false-discovery rate (FDR) perform in genome-wide association and fine mapping analyses?
Conclusion
FDR is effective for genome-wide association studies but may not be reliable for fine mapping due to a high proportion of significant results.
Supporting Evidence
- FDR performed well in genome-wide analysis with good localization of positive results.
- In fine mapping, all tested methods performed poorly, producing a high proportion of significant results.
- FDR was empirically more conservative than expected in genome-wide association studies.
Takeaway
This study looked at how well a method called FDR works when finding important genetic markers. It found that while FDR is good for big studies, it can give too many false positives in smaller, detailed studies.
Methodology
The study analyzed simulated data from 1500 families with rheumatoid arthritis, examining 9187 SNPs in genome-wide and 2094 SNPs in fine mapping.
Potential Biases
The study suggests that the distribution of p-values may bias FDR performance in fine mapping.
Limitations
The performance of FDR in fine mapping was poor, leading to a high proportion of significant results.
Participant Demographics
1500 families with an affected sibling pair and 2000 unaffected controls.
Statistical Information
P-Value
p < 10^-300
Statistical Significance
p<0.05
Want to read the original?
Access the complete publication on the publisher's website