Related Experiment Videos
Haplotype interaction analysis of unlinked regions.
Tim Becker1, Johannes Schumacher, Sven Cichon
1Institute for Medical Biometry, Informatics and Epidemiology, University of Bonn, Bonn, Germany. becker@imbie.meb.uni-bonn.de
Genetic Epidemiology
|October 22, 2005
Summary
This study introduces a novel statistical method for analyzing complex genetic diseases by examining gene-gene interactions across multiple genomic regions. The approach enhances power for detecting disease associations, even with unlinked regions, and is implemented in FAMHAP.
Area of Science:
- Genetics and Bioinformatics
- Statistical Genomics
- Computational Biology
Background:
- Genetically complex diseases arise from interactions between environmental factors and multiple genes.
- Analyzing these complex interactions requires statistical methods capable of considering numerous genomic regions simultaneously.
- High-dimensional statistical tests for gene-gene interactions present significant analytical challenges, particularly regarding multiple testing.
Purpose of the Study:
- To develop and validate a statistical method for case-control studies that analyzes gene-gene interactions across multiple unlinked genomic regions.
- To address the challenge of multiple testing in high-dimensional genetic association studies.
- To improve the power of detecting genetic associations by considering haplotype interactions.
Main Methods:
- A novel method for analyzing case-control studies using multi-SNP data without phase information.
- Incorporation of haplotype interactions between different unlinked genomic regions.
- Application of the minP approach with Monte-Carlo simulations for multiple testing correction.
- Global hypothesis testing for any significant haplotype interaction effect.
Main Results:
- Simulation studies confirmed the validity and increased power of the proposed method compared to single-region analyses for two-locus disease models.
- The method demonstrated robustness, with minimal power loss when one region was not involved in disease etiology.
- Successful application to a real case-control dataset identified significant gene-gene interactions missed by classical analysis, even after multiple testing correction.
Conclusions:
- The developed statistical method effectively analyzes gene-gene interactions in complex diseases using multi-SNP data.
- The approach provides a powerful tool for genetic association studies, particularly for identifying interactions between unlinked genomic regions.
- The FAMHAP software implementation facilitates the application of this advanced haplotype interaction analysis.