Skip to main content

Table 11 Power of FORCE detection method, impact of various sources of noise and combinations of them for the 6 epistatic models

From: Filter-free exhaustive odds ratio-based genome-wide interaction approach pinpoints evidence for interaction in the HLA region in psoriasis

Type of noise Model 1 a Model 2 Model 3 Model 4 Model 5 Model 6
No noise 1 0.97 0.96 1 0.93 0.99
Genotype errors (GE) 0.99 0.99 0.97 1 0.95 0.99
Phenocopy (PC) 0.08 0.55 0.13 0.33 0.23 0.37
Misssing data (MS) 0.99 0.96 0.95 1 0.93 0.99
GE + PC 0.05 0.62 0.18 0.3 0.31 0.35
GE + MS 0.95 0.98 0.96 1 0.91 0.99
PC + MS 0.06 0.52 0.21 0.31 0.21 0.26
GE + PC + MS 0.09 0.55 0.21 0.46 0.13 0.35
  both first either both first either both first either both first either both first either both first either
Genetic heterogeneity (GH) 0.01 0.08 0.14 0.38 0.6 0.82 0 0.19 0.34 0.2 0.39 0.62 0 0.21 0.38 0.1 0.34 0.57
GH + GE 0.01 0.09 0.16 0.34 0.6 0.85 0.03 0.18 0.32 0.17 0.4 0.62 0.04 0.23 0.41 0.14 0.31 0.50
GH + PC 0 0.015 0.03 0.02 0.09 0.16 0 0.01 0.02 0 0.02 0.04 0.01 0.035 0.06 0.01 0.04 0.07
GH + MS 0.01 0.04 0.07 0.37 0.57 0.77 0.02 0.145 0.27 0.18 0.385 0.59 0.03 0.19 0.35 0.07 0.28 0.50
GH + GE + PC 0 0.01 0.02 0.03 0.105 0.18 0 0.02 0.04 0 0.025 0.05 0 0.025 0.05 0 0.03 0.06
GH + GE + MS 0 0.05 0.1 0.33 0.665 0.80 0.02 0.155 0.29 0.13 0.385 0.64 0.03 0.23 0.43 0.13 0.305 0.48
GH + PC + MS 0 0.005 0.01 0.01 0.095 0.18 0 0.025 0.05 0 0.04 0.08 0 0.005 0.01 0 0.035 0.07
GH + GE + PC + MS 0 0.015 0.03 0.01 0.08 0.15 0 0.015 0.03 0 0.035 0.07 0 0.015 0.03 0 0.03 0.06
  1. GE: Genotyping errors – GH: Genetic heterogeneity – MS: Missing data – PC: Phenocopy. In case of GH, power is calculated in 3 different ways as the proportion of datasets in which both, the first or either of the interacting pairs are detected. aIn bold, power higher than 50%.