Table 2.
Generalization testing results from a set of analyses based on a HCHS/SOL GWAS as the discovery study, and GLGC GWAS as the follow-up study. For each selection rule we report the number of SNPs selected for follow-up testing, and the number of loci containing these SNPs. For combinations of selection rules and multiple testing adjustment method we report the number of generalized loci, and the number of generalized loci that did not contain any SNP with p-value< 5 × 10−8 in the GLGC GWAS.
| Selection rule | selected SNPs | loci | adjustment | gen SNPs | gen loci | # loci not sig Willer |
|---|---|---|---|---|---|---|
| Selection rule 1 - one-sided | 1,662 | 89 | FDRg r-values | 1,352 | 27 | 5 |
| Selection rule 1 - two-sided | 1,208 | 51 | FDRg r-values | 1,076 | 21 | 1 |
| 10−6 | 742 | 18 | FDRg r-values | 706 | 17 | 0 |
| 10−6 | 742 | 18 | FWERg r-values | 583 | 17 | 0 |
| Selection rule 2 - one-sided | 627 | 17 | FWERg r-values | 583 | 17 | 0 |
| Selection rule 2 - two-sided | 574 | 16 | FWERg r-values | 538 | 16 | 0 |
| 5 × 10−8 | 546 | 15 | FWERg r-values | 514 | 15 | 0 |