Related Experiment Videos
Cox proportional hazards survival regression in haplotype-based association analysis using the Stochastic-EM
1INSERM U525, 75634 Paris, France. david.tregouet@chups.jussieu.fr
European Journal of Human Genetics : EJHG
|July 9, 2004
Summary
This study introduces a novel method for analyzing genetic association studies with survival outcomes. The Stochastic-EM algorithm is adapted for haplotype effect estimation in censored data, enhancing parameter estimation efficiency.
Area of Science:
- Genetics
- Biostatistics
- Computational Biology
Background:
- Haplotype information from genotypes is crucial for understanding complex traits in association studies.
- Simultaneous estimation of haplotype frequencies and effects improves parameter estimation efficiency.
- Existing models primarily address binary or quantitative phenotypes, lacking survival outcome analysis.
Purpose of the Study:
- To extend haplotype-based association analysis to survival outcomes.
- To apply the Stochastic-EM (SEM) algorithm for estimating haplotype effects in censored data.
- To implement this novel model in the THESIAS software.
Main Methods:
- Utilizing the Stochastic-EM (SEM) algorithm.
- Applying a standard Cox proportional hazards formulation for censored data analysis.
- Implementing the model within the THESIAS software package.
Main Results:
- The SEM algorithm is successfully adapted for haplotype effect estimation in survival analysis.
- The developed model provides an efficient method for parameter estimation with censored data.
- The THESIAS software facilitates the application of this method.
Conclusions:
- Haplotype-based association analysis can be effectively applied to survival outcomes.
- The SEM algorithm offers a robust approach for estimating haplotype effects in censored data.
- The THESIAS software provides a valuable tool for researchers in this area.