Table 5. Observed versus expected number of significant associations, by trait and population.
Trait | Race/Ethnicity | # of Total Tests of Association* | # of Observed Significant Associations | # of Expected Significant Associations† | P-value‡ |
HDL-C | AA | 23 | 11 | 17.3 | 0.01 |
AI | 20 | 9 | 14.4 | 0.01 | |
MA/H | 23 | 13 | 13.8 | 0.83 | |
LDL-C | AA | 18 | 11 | 14.7 | 0.03 |
AI | 14 | 10 | 11.9 | 0.15 | |
MA/H | 18 | 10 | 10.6 | 0.81 | |
ln(TG) | AA | 14 | 8 | 11.9 | 0.01 |
AI | 13 | 10 | 8.4 | 0.56 | |
MA/H | 14 | 12 | 10.4 | 0.54 |
The expected number of significant associations was based on power calculations assuming an additive genetic model and liberal significance threshold (0.05) in each racial/ethnic group for each test of association. We further assumed the observed genetic effect size (beta) from PAGE European Americans and the observed allele frequency, sample sizes, and trait mean/standard deviations from each non-European American population.
*Only includes associations that replicated in EA.
†Based on the additive power of all loci.
‡One-sample binomial test.
African Americans (AA); American Indians (AI); Mexican Americans/Hispanics (MA/H); number (#).