Simultaneous Analysis of All SNPs in Genome-Wide Association Studies
Author Information
Author(s): Clive J. Hoggart, John C. Whittaker, Maria De Iorio, David J. Balding
Primary Institution: Imperial College, London
Hypothesis
Can simultaneous analysis of all SNPs improve the identification of causal variants in complex diseases?
Conclusion
The simultaneous analysis of SNPs improves the identification of causal variants and reduces false positives compared to traditional single-SNP methods.
Supporting Evidence
- Simultaneous analysis of SNPs leads to improved SNP selection compared to single-SNP tests.
- The method can accommodate both quantitative and case-control phenotypes.
- Using simulated data, the method demonstrated lower false positive rates than traditional methods.
- The analysis can be performed efficiently on standard desktop workstations.
Takeaway
This study shows that looking at all genetic markers at once helps scientists find the ones that really affect diseases better than checking them one by one.
Methodology
A Bayesian-inspired penalised maximum likelihood approach was used to analyze SNPs simultaneously for their association with disease outcomes.
Potential Biases
Potential bias due to the reliance on prior distributions and the selection of SNPs based on their association with disease.
Limitations
The method may overlook SNPs with strong marginal effects if other SNPs better explain the disease risk.
Participant Demographics
The study involved 1000 cases and 1000 controls from a population of 10,000 individuals.
Statistical Information
P-Value
0.011
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website