Fine mapping versus replication in whole-genome association studies.

Area of Science:

• Genetics
• Biostatistics
• Genomic Epidemiology

Background:

• Replication studies in genetic association research frequently report different findings than primary studies, raising questions about genuine associations versus false positives.
• Understanding replication outcomes is crucial for genomewide association (GWA) studies, which identify numerous potential associations requiring external validation.
• Current replication practices often test variants beyond those initially identified, blurring the lines between replication and fine-mapping.

Purpose of the Study:

• To compare the efficacy of two marker-selection strategies in replication studies: an 'exact' approach (original markers only) and a 'local' approach (original and nearby markers).
• To evaluate the impact of marker selection on the probability of successful gene or region replication in genetic association studies.

Main Methods:

• Simulated replication scenarios comparing an 'exact' marker selection strategy against a 'local' marker selection strategy.
• Analysis of the probability of replicating a genetic association under varying conditions, including linkage disequilibrium and study power.

Main Results:

• The 'exact' approach, using only initially identified variants, is the most powerful and efficient replication strategy in most scenarios.
• The 'local' approach is advantageous only when replicating findings in a region with high intermarker linkage disequilibrium and weak initial associations.
• Replication with different markers can occur frequently, even with true associations, especially in low-powered studies with increased numbers of tested variants.

Conclusions:

• The choice of marker selection strategy significantly impacts the success and interpretation of genetic association replication studies.
• Clear distinction between fine-mapping and replication is essential for accurate interpretation of GWA study results.
• The 'exact' replication strategy is recommended for most GWA replication efforts to maximize power and minimize false positives.