Using Principal-Component Analysis for Genetic Association Testing
Author Information
Author(s): Oh Sohee, Park Taesung
Primary Institution: Seoul National University
Hypothesis
Can principal-component association tests provide better performance than haplotype-based association tests in genetic studies?
Conclusion
The principal-component test outperformed the haplotype-based test by having smaller AIC values and slightly greater power.
Supporting Evidence
- The principal-component test showed smaller AIC values compared to the haplotype-based test.
- The PC score test demonstrated greater power in detecting associations with the disease.
- Using PC scores reduces the number of parameters needed in logistic regression.
Takeaway
This study shows that using a new method called principal-component analysis can help scientists find genetic links to diseases more easily than older methods.
Methodology
The study used logistic regression to compare the performance of principal-component tests and haplotype-based tests on simulated genetic data.
Limitations
The interpretation of principal-component scores is not straightforward, and the biological meaning of these scores is difficult to ascertain.
Participant Demographics
The study included 1500 individuals from affected sibling pairs and 500 controls.
Statistical Information
Statistical Significance
p<0.05
Want to read the original?
Access the complete publication on the publisher's website