Model Selection for Identifying Causal Variants in SNPs
Author Information
Author(s): Uh Hae-Won, Mertens Bart JA, Jan van der Wijk Henk, Putter Hein, van Houwelingen Hans C, Houwing-Duistermaat Jeanine J
Primary Institution: Leiden University Medical Center
Hypothesis
Can we effectively identify causal variants from a dense panel of SNPs using logistic regression methods?
Conclusion
The Bayesian variable selection method outperformed other methods in identifying causal SNPs and improving prediction performance.
Supporting Evidence
- The Bayesian method identified only two SNPs, demonstrating its parsimony.
- Average prediction error was lowest for the Bayesian method compared to others.
- Sensitivity and specificity of the Bayesian model were found to be 0.82.
Takeaway
The study looked at how to find important genetic markers in a group of similar genes, and found that a special method called Bayesian worked best.
Methodology
The study used logistic regression with stepwise selection, lasso penalty, and Bayesian methods to analyze SNP data.
Limitations
The validation set size was small, which may affect the robustness of the findings.
Participant Demographics
200 cases and 200 controls based on rheumatoid arthritis status.
Statistical Information
P-Value
0.001
Statistical Significance
p<0.001
Want to read the original?
Access the complete publication on the publisher's website