A comparison of tagging methods and their tagging space

Area of Science:

• Genetics
• Bioinformatics
• Statistical genomics

Background:

• Single-nucleotide polymorphism (SNP) tagging is crucial for reducing genotyping costs in genetic association studies.
• Various tagging methods aim to minimize markers while retaining statistical power for ungenotyped SNPs.
• Understanding the performance, overlap, and differences between these methods is essential.

Purpose of the Study:

• To compare the performance and concordance of three widely used SNP tagging methods.
• To evaluate how different tagging algorithms perform across various SNP frequencies and settings.
• To assess the overlap in selected tagging SNPs between distinct methods.

Main Methods:

• Comparison of a haplotype r2-based, a pairwise r2-based, and a haplotype diversity-based tagging method.
• Evaluation using ENCODE regions genotyped on HapMap CEPH individuals.
• Assessment of tagging efficiency (genotyped markers/tagging SNPs) and effectiveness (detection of hidden SNPs).

Main Results:

• Tagging effectiveness was generally lower for rare SNPs compared to common SNPs across all methods.
• Including rare SNPs in tagging schemes improved detection of rare hidden SNPs but increased genotyping costs.
• High concordance (>90%) was observed in detected hidden SNPs between methods at moderate tagging efficiency, reaching 100% with lower efficiency.

Conclusions:

• Different SNP tagging methods exhibit significant concordance in identifying effective tagging SNPs.
• The choice of tagging method may have less impact on overall study power than anticipated due to high overlap.
• Optimizing tagging strategies requires balancing efficiency, effectiveness, and the inclusion of rare variants.