Table 3. SNPs found significant across all phenotypes after correction for FWER (Holm) and FDR (Benjamini-Hochberg), both at a threshold of 0.05.
| Including duplicates |
Unique SNPs |
|||
|---|---|---|---|---|
| Method | Holm | BH | Holm | BH |
| R-LM | 389 | 2662 | 99 | 1200 |
| R-Q | 249 | 1024 | 106 | 415 |
| EMMAX-K | 235 | 353 | 31 | 57 |
| EMMAX-QK | 68 | 97 | 14 | 29 |
| R-LA | 431 | 2154 | 220 | 891 |
| R-LAQ | 191 | 1261 | 107 | 723 |
| EMMAX-KLA | 637 | 1477 | 65 | 171 |
| EMMAX-KLAQ | 90 | 144 | 17 | 44 |
| BEST | 162 | 359 | 63 | 177 |
The multiple testing corrections were applied to each condition separately, not to all 201 phenotypes all together. SNPs that are found to be significant in multiple phenotypes are included multiple times under the “Including duplicates” but only once under the “Unique SNPs.” BH, Benjamini-Hochberg.