Reducing Selection Bias in Genetic Effect Estimation
Author Information
Author(s): Wu Long Yang, Lee Sophia SF, Shi Haijiang Steven, Sun Lei, Bull Shelley B
Primary Institution: Samuel Lunenfeld Research Institute, Mount Sinai Hospital
Hypothesis
Can bootstrap-based resampling methods effectively reduce selection bias in genetic effect estimation during genome-wide scans?
Conclusion
The study found that bootstrap-based estimators effectively reduce selection bias in genetic effect estimation for both true and false positives.
Supporting Evidence
- The bootstrap-based estimators reduced the upward selection bias in genetic effect estimation for both microsatellite and SNP based linkage analysis.
- The shrinkage estimator is recommended when the power to detect the disease locus is low.
- The weighted estimator is recommended when the power is higher.
Takeaway
This study shows that using special methods can help scientists get better estimates of how genes affect diseases, especially when the initial results might be misleading.
Methodology
The study used simulated data and conducted multipoint analyses with bootstrap resampling methods to evaluate genetic effect estimates.
Potential Biases
The study acknowledges that upward selection bias can occur due to strict significance criteria.
Limitations
The study focused only on the most significant locus in genome scans and did not consider other loci that exceeded significance criteria.
Participant Demographics
The study involved four populations: Aipotu, Danacaa, Karangar, and New York, with varying family structures.
Statistical Information
P-Value
2.2 × 10-5
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website