Identifying susceptibility genes using linkage and linkage disequilibrium analysis in large pedigrees

Area of Science:

• Genetics
• Statistical genetics
• Genomic association studies

Background:

• Linkage analysis and linkage disequilibrium (LD) are crucial for mapping genes underlying complex diseases.
• Identifying disease-associated markers requires robust statistical approaches.

Purpose of the Study:

• To compare the power of two distinct strategies for identifying disease-associated markers.
• To evaluate the utility of Simes' test for combining p-values from LD tests on correlated markers.

Main Methods:

• Applied two primary approaches: 1) linkage analysis followed by LD tests within linked regions, and 2) direct LD tests on all markers.
• Utilized the Genetic Analysis Workshop 12 (GAW12) problem 2 isolated population dataset across multiple replicates.
• Investigated the use of Simes' test to aggregate p-values from LD tests for adjacent markers.

Main Results:

• The approach of testing all markers for LD with disease demonstrated higher statistical power compared to the sequential linkage and LD testing method.
• The benefit of employing Simes' test to combine p-values for correlated markers in LD tests was found to be marginal.
• Results were initially reported for a single replicate, with overall method power assessed using all available replicates.

Conclusions:

• Directly testing all markers for linkage disequilibrium with disease is a more powerful strategy for complex disease gene mapping than a stepwise linkage-then-LD approach.
• The application of Simes' test for combining p-values in LD analyses of correlated markers offers limited added value.
• These findings inform the selection of optimal statistical methods for genetic association studies in complex diseases.