Related Experiment Videos
Genomic control for association studies: a semiparametric test to detect excess-haplotype sharing
B Devlin1, K Roeder, L Wasserman
1Department of Psychiatry, University of Pittsburgh, Pittsburgh, PA 15213, USA.
Biostatistics (Oxford, England)
|August 23, 2003
Summary
This study introduces a new statistical test to identify disease genes by analyzing shared genetic marker patterns (haplotypes) among affected individuals. The method detects increased haplotype sharing near disease mutations, aiding genetic disease research.
Area of Science:
- Genetics
- Statistical genetics
- Bioinformatics
Background:
- Individuals sharing a disease mutation often inherit similar adjacent genetic markers (haplotypes) from a common ancestor.
- Haplotype sharing is expected to be more pronounced near disease genes compared to other genomic regions.
Purpose of the Study:
- To develop a semiparametric statistical test for detecting haplotype sharing.
- To identify genetic markers associated with disease mutations by quantifying haplotype clustering.
Main Methods:
- Developed a model assuming a known ancestral haplotype to measure haplotype sharing unambiguously.
- Estimated unknown distributions non-parametrically and modeled marker overlap near disease genes using a mixture distribution.
- Introduced a pairwise haplotype overlap measure and a score test when the ancestral haplotype is unknown.
Main Results:
- The test effectively distinguishes between regions with and without disease genes based on haplotype sharing patterns.
- The developed statistical framework provides a method for localizing disease genes.
Conclusions:
- The semiparametric haplotype-sharing test is a valuable tool for genetic disease gene mapping.
- This approach enhances the ability to detect disease-associated genomic regions through statistical analysis of haplotype data.