Power to detect risk alleles using genome-wide tag SNP panels

Abstract

Advances in high-throughput genotyping and the International HapMap Project have enabled association studies at the whole-genome level. We have constructed whole-genome genotyping panels of over 550,000 (HumanHap550) and 650,000 (HumanHap650Y) SNP loci by choosing tag SNPs from all populations genotyped by the International HapMap Project. These panels also contain additional SNP content in regions that have historically been overrepresented in diseases, such as nonsynonymous sites, the MHC region, copy number variant regions and mitochondrial DNA. We estimate that the tag SNP loci in these panels cover the majority of all common variation in the genome as measured by coverage of both all common HapMap SNPs and an independent set of SNPs derived from complete resequencing of genes obtained from SeattleSNPs. We also estimate that, given a sample size of 1,000 cases and 1,000 controls, these panels have the power to detect single disease loci of moderate risk (lambda approximately 1.8-2.0). Relative risks as low as lambda approximately 1.1-1.3 can be detected using 10,000 cases and 10,000 controls depending on the sample population and disease model. If multiple loci are involved, the power increases significantly to detect at least one locus such that relative risks 20%-35% lower can be detected with 80% power if between two and four independent loci are involved. Although our SNP selection was based on HapMap data, which is a subset of all common SNPs, these panels effectively capture the majority of all common variation and provide high power to detect risk alleles that are not represented in the HapMap data.

Figure 1. Coverage of the HapMap (A) or SeattleSNPs (B) Datasets As Measured by the Proportion of Common Variation in Each Dataset That Is Captured by a SNP in Either HumanHap550 (CEU; CHB + JPT) or HumanHap650Y (YRI) at Various r ² Thresholds

Lines show coverage for the CEU (red), CHB + JPT (green), and YRI (blue) populations.

Figure 2. Total Power to Detect a Single Common Risk Allele in the HapMap Data Using HumanHap550 for CEU (A) and CHB + JPT (B) and HumanHap650Y for YRI (C) in a Study Size of 1,000 cases and 1,000 Controls under a Multiplicative Disease Model

Power is calculated for a 5% false discovery rate after a Bonferroni correction for multiple testing (red line). The solid black line represents the power to detect a risk allele if all the common HapMap SNPs in each respective population are genotyped using the same significance cutoff. Also shown are the powers for different frequency ranges as dashed lines, where the lower line indicates the power for risk allele frequencies from 5%–10% and the upper line indicates the power for risk allele frequencies from 40%–50%.

Figure 3. The Minimum Risk Detectable at 80% Power with p ≤ 0.05 after a Bonferroni Correction for Multiple Testing (χ² ≥ 28.5687) for Various Sample Sizes under Multiplicative (Red) and Additive (Blue) Models

Power is calculated as the ability to detect a single common risk allele in the HapMap data in a whole genome association study within CEU samples.

Figure 4. Total Power to Detect a Single Common Risk Allele outside of the HapMap Data Using 1,000 cases and 1,000 Controls under a Multiplicative Model

The power is estimated using the common SNPs in 68 resequenced SeatleSNPs genes that were not characterized in the HapMap Project for (A) 23 CEU samples using HumanHap550 and (B) 24 YRI samples using HumanHap650Y. Power is calculated for a 5% false discovery rate after a Bonferroni correction for multiple testing (red line). The solid black line shows the power to detect a common risk allele in the HapMap data (i.e., solid red lines in Figure 2).

Figure 5. Power to Detect At Least One Risk Allele in CEU (HumanHap550) and YRI (HumanHap650Y) in 1,000 Cases and 1,000 Controls under a Multiplicative Model

Curves represent the cases where there is a single risk locus (red), two independent loci (blue), or four independent loci (green). The red lines are the single risk loci as shown in Figure 4. Under the multiple loci cases, the relative risk represents the minimum risk for all risk alleles and the corresponding power represents a lower bound on the power.