Relaxed significance criteria for linkage analysis

Area of Science:

• Genetics
• Statistical genetics
• Genomic analysis

Background:

• Linkage analysis identifies gene loci associated with traits.
• Traditional genomewise error rate (GWER) controls false positives strictly.
• Complex traits often involve multiple quantitative trait loci (QTL).

Purpose of the Study:

• Evaluate challenges of relaxed significance criteria like FDR in linkage analysis.
• Assess the suitability of FDR for single-trait linkage studies.
• Propose a novel, more applicable significance criterion.

Main Methods:

• Characterized problems with FDR in single-trait linkage analysis.
• Proposed a generalized genomewise error rate (GWERk) criterion.
• Compared GWERk to traditional GWER and FDR.

Main Results:

• Identified several problems with applying FDR to single-trait linkage analysis.
• Demonstrated that FDR's general applicability in this context is questionable.
• Showcased GWERk as a viable alternative offering a relaxed balance of true and false positives.

Conclusions:

• FDR is not generally suitable for declaring significant linkages in single-trait analysis.
• GWERk provides a more liberal significance threshold than GWER.
• GWERk balances true and false positives effectively without added computational burden or assumptions.