Estimating Effects of Rare Haplotype on Failure Time
Author Information
Author(s): Olga W. Souverein, Aeilko H. Zwinderman, Wouter J. Jukema, Michael W. T. Tanck
Primary Institution: Academic Medical Center, University of Amsterdam
Hypothesis
Can a penalized Cox proportional hazards regression model effectively estimate the effects of rare haplotypes on failure time in unrelated individuals?
Conclusion
The penalized log-likelihood approach produces valid results for studying the association between rare haplotypes and failure time.
Supporting Evidence
- The method allows for adjustment for environmental covariates and haplotype-environment interactions.
- Simulations showed that the penalized log-likelihood approach can stabilize estimates of rare haplotypes.
- Significant interactions were found between certain SNPs affecting target vessel revascularization risk.
Takeaway
This study shows a new way to understand how rare genetic variations affect health by using a special math method.
Methodology
The study used a penalized Cox proportional hazards regression model with an EM algorithm to estimate haplotype effects.
Potential Biases
Potential bias due to the assumption that similar haplotypes show similar effects.
Limitations
The method may not be robust to violations of the Hardy-Weinberg assumption.
Participant Demographics
Patients with cardiovascular disease treated with PTCA, with a focus on four SNPs in the IL10 gene.
Statistical Information
P-Value
0.003 for interactions between SNPs
Confidence Interval
95% CI: 1.00–1.46 for IL10 -2849G > A
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website