Table 2. Direction and significance (P<0.05) of effect estimates for the Add Health analytic sample, compared with published results for 41 SNPsa.
N | Of 41 SNPs, the no. in same direction as published resultsb | P-value | No. of SNPs in same direction as published result and P<0.05c | P-value | |
---|---|---|---|---|---|
European American adolescents | 5103 | 35 | 2.04E-06 | 18 | 2.37E-13 |
African American adolescents | 1748 | 25 | 0.05 | 8 | 6.87E-04 |
Hispanic American adolescents | 1304 | 30 | 0.001 | 3 | 0.2 |
Asian American adolescents | 439 | 24 | 0.07 | 5 | 0.04 |
Meta-analyzed sample, all adolescents | 8594 | 36 | 3.41E-07 | 21 | 4.60E-17 |
Abbreviation: SNP, single-nucleotide polymorphism.
β effect estimate for SNP–BMI associations.
Direction of effect estimates and nominal significance at P<0.05 is compared with direction of effect estimates and nominal significance in published results with European American adults,6, 7, 8, 9, 30, 31 for 41 SNPs.
The number of SNPs for which the direction of effect is concordant between published results and results in Add Health adolescent in the combined sample and by race/ethnicity. The P-value reported is one-tailed, based on a binomial draw with null expectation P=0.5.
The number of SNPs for which the P-value for SNP–BMI association in results in Add Health adolescent in the combined sample and by race/ethnicity directionally consistent and significant at P<−0.05. The P-value reported is one-tailed, based on a binomial draw with null expectation P=0.05.