Comparing Methods for Analyzing Gene-Environment Interactions in Rheumatoid Arthritis
Author Information
Author(s): Dempfle Astrid, Hein Rebecca, Beckmann Lars, Scherag André, Nguyen Thuy Trang, Schäfer Helmut, Chang-Claude Jenny
Primary Institution: Institute of Medical Biometry and Epidemiology, Philipps-University Marburg
Hypothesis
Can haplotype-based methods improve the detection of gene × environment interactions compared to single- and multilocus association methods?
Conclusion
Haplotype-based methods showed high power for detecting gene × sex interactions in rheumatoid arthritis, while stepwise methods performed poorly due to strong linkage disequilibrium.
Supporting Evidence
- Haplotype-based analyses had high power to detect gene × sex interactions.
- Stepwise conditional logistic regression performed poorly due to collinearity among predictors.
- The study utilized both simulated and real data to compare the effectiveness of different methods.
Takeaway
This study looked at how different methods can help find out if genes and environmental factors, like smoking, work together to affect rheumatoid arthritis risk.
Methodology
The study used logistic regression and haplotype-based methods to analyze simulated and real data on rheumatoid arthritis.
Potential Biases
The control sample included a higher percentage of females and smokers than the case sample, leading to potential bias in estimating main effects.
Limitations
The control sample was not representative of the population regarding sex and smoking distribution, which may bias the results.
Participant Demographics
The sample included 665 unrelated RA cases and 336 unrelated unaffected controls, with a female-to-male ratio of approximately 2:1 in the matched case-control pairs.
Statistical Information
P-Value
<0.001
Statistical Significance
p<0.05
Want to read the original?
Access the complete publication on the publisher's website