Understanding Gene-Environment Interactions in Genetic Studies
Author Information
Author(s): Arend Voorman, Thomas Lumley, Barbara McKnight, Kenneth Rice
Primary Institution: Department of Biostatistics, University of Washington, Seattle, Washington, United States of America
Hypothesis
How do population substructure and variability affect the interpretation of gene-environment interaction studies?
Conclusion
Using model-robust standard errors can prevent misleading conclusions about population substructure in gene-environment interaction studies.
Supporting Evidence
- Model-robust standard errors provide better estimates of variability than model-based standard errors.
- Mis-specification of the mean model can lead to inflated test statistics in gene-environment studies.
- Simulation results indicate that model-robust methods yield well-calibrated p-values.
Takeaway
This study shows that using the wrong methods can make it look like there are differences in genes when there aren't any. Better methods can help us see the truth.
Methodology
The study used simulations to compare model-based and model-robust standard errors in gene-environment interaction analyses.
Potential Biases
Potential bias from using model-based standard errors in the presence of model mis-specification.
Limitations
The study primarily focuses on simulation results, which may not fully capture real-world complexities.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website