Table 1 Number of significant interactions detected in Stage 2 of the simultaneous and the conditional designs by logistic regression modeling [Eq. (3) and (4)].
From: Two-stage strategies to detect gene × gene interactions in case-control data
Full model without covariates [Eq. (3)] | Additive effects with covariates [Eq. (4)] | ||||
|---|---|---|---|---|---|
i aa | i ad | i da | i dd | i aa | |
Simultaneous design p < 0.05 | 863,970 tests | 399,171 tests | |||
<1,000 kb | 46 | 0 | 0 | 1 | 206 |
>1,000 kb | 11 | 0 | 0 | 0 | 70 |
Simultaneous design p < 0.10 | 925,480 tests | 869,221 tests | |||
<1,000 kb | 46 | 0 | 0 | 1 | 209 |
>1,000 kb | 11 | 0 | 0 | 0 | 93 |
Conditional design | 4,069,398 tests | 3,656,028 tests | |||
<1,000 kb | 49 | 0 | 0 | 0 | 353 |
>1,000 kb | 75 | 0 | 0 | 4 | 617 |