Robust Ranks of True Associations in Genome-Wide Studies
Author Information
Author(s): Zheng Gang, Joo Jungnam, Lin Jing-Ping, Stylianou Mario, Waclawiw Myron A, Geller Nancy L
Primary Institution: National Heart, Lung and Blood Institute
Hypothesis
Can robust ranking methods improve the identification of true genetic associations in genome-wide association studies?
Conclusion
Using robust ranking methods like min p and MERT can enhance the power to identify true genetic associations when the underlying genetic model is unknown.
Supporting Evidence
- Robust ranking methods can lead to higher power in identifying true associations.
- Using a single trend test may not be reliable when the true genetic model is unknown.
- Min p and MERT methods showed better performance in ranking SNPs compared to single tests.
Takeaway
This study shows that using better methods to rank genetic markers can help find the right ones that are linked to diseases, even when we don't know the exact genetic rules.
Methodology
The study employed a two-stage analysis for genetic markers, using trend tests and robust tests to rank SNPs based on their association with diseases.
Limitations
The study relies on simulations, which may not fully capture real-world complexities of genetic associations.
Participant Demographics
500 cases and 500 controls were simulated in the independent study.
Statistical Information
P-Value
p<0.05
Statistical Significance
p<0.05
Want to read the original?
Access the complete publication on the publisher's website