Impact of Weighting on Haseman-Elston Regression for Genetic Analysis
Author Information
Author(s): Franke Daniel, Kleensang André, Elston Robert C, Ziegler Andreas
Primary Institution: Institut für Medizinische Biometrie und Statistik, Universität zu Lübeck
Hypothesis
Does weighting the Haseman-Elston regression by marker informativity affect type I error rates?
Conclusion
Weighting Haseman-Elston regression models by informativity can inflate type I error rates.
Supporting Evidence
- Weighting schemes can lead to an increase in significant findings.
- The classical Haseman-Elston method maintains its nominal significance level.
- Empirical p-values are recommended for weighted methods to avoid inflated error rates.
Takeaway
This study looked at how adjusting for certain genetic information can change the results of genetic tests, and found that it can sometimes lead to incorrect conclusions.
Methodology
The study used simulated data to compare classical and weighted Haseman-Elston regression methods across 100 replicates.
Potential Biases
Potential bias due to inflated type I error rates when using weighted methods without empirical p-values.
Limitations
The study used simulated data, which may not fully represent real-world scenarios.
Participant Demographics
Data from the Aipotu population was used, but specific demographics were not detailed.
Statistical Information
P-Value
3.3 × 10-7
Confidence Interval
95% CI: 0.0546–0.0823
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website