Comparing Genetic Association Test Strategies
Author Information
Author(s): Tian Xin, Joo Jungnam, Wu Colin O, Lin Jing-Ping
Primary Institution: Office of Biostatistics Research, National Heart, Lung and Blood Institute
Hypothesis
How do different test strategies for genetic association in case-control studies using family data compare in terms of power?
Conclusion
All three tested methods for genetic association performed comparably well in the simulation and application to the candidate gene data.
Supporting Evidence
- The study compared the power of three genetic association test strategies.
- Simulation results indicated that the GEE and MO methods had good power ranging from 73% to 84%.
- All three methods identified the same significant markers associated with rheumatoid arthritis.
Takeaway
The study looked at different ways to test if genes are linked to diseases using family data, and found that all methods worked well.
Methodology
The study compared three methods: the Cochran-Armitage trend test, generalized estimating equations, and multiple outputation through simulation and application to genetic data.
Potential Biases
Potential for increased false positive rates if correlation among related individuals is ignored.
Limitations
The study's methods are sensitive to population stratification and may require large family samples for certain tests.
Participant Demographics
The study involved 200 cases from 100 families with affected siblings and 200 unrelated controls.
Statistical Information
P-Value
<10-6
Want to read the original?
Access the complete publication on the publisher's website