Imputation-Based Analysis of Association Studies
Author Information
Author(s): Servin Bertrand, Stephens Matthew
Primary Institution: Department of Statistics, University of Washington
Hypothesis
Can untyped genetic variants be effectively tested for association with phenotypes using imputation methods?
Conclusion
The study presents a new framework that improves the detection of associations between genetic variants and phenotypes by effectively imputing untyped SNPs.
Supporting Evidence
- The new framework allows for more effective testing of untyped SNPs.
- Imputation methods provide greater power to detect associations.
- Results closely approximate those obtained by genotyping all SNPs.
Takeaway
This study shows a way to guess missing genetic information to better understand how genes affect traits, making it easier to find important genetic links.
Methodology
The study uses a Bayesian regression approach to impute genotypes at untyped SNPs and assess their association with phenotypes.
Potential Biases
Potential biases may arise from the assumptions made in the Bayesian framework and the choice of priors.
Limitations
The methods may not perform as well for rare variants and rely on the accuracy of imputation.
Participant Demographics
Participants were of European descent, including parents from 32 trios and 425 patients.
Statistical Information
P-Value
0.006
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website