Skip to main content

Table 3 TDR for detecting genotype effects in the presence of a continuous covariate

From: Application of two machine learning algorithms to genetic association studies in the presence of covariates

 

Approach

MODEL

Include Z

Residualize by Z

Ignore Z

MODEL 1: ADDITIVE

   

β/σ = γ/σ = 0.5

0.969/0.546

0.993/0.999

0.965/0.995

MODEL 2: ADDITIVE WITH CONFOUNDING

   

β/σ = γ/σ = 0.5

0.922/0.729

0.942/0.731

0.116/0.740

MODEL 3: ADDITIVE WITH EFFECT MEDIATION

   

β/σ = 0.5, γ/σ = 2

0.829/0.703

0.133/0.033

0.510/0.628

MODEL 4: INTERACTION

   

β/σ = γ/σ = ζ/σ = 0.5

0.999/0.909

0.995/0.998

0.930/0.984

MODEL 5: INTERACTION, NO MAIN EFFECTS

   

ζ/σ = 0.5

0.898/0.963

0.137/0.033

0.137/0.036

MODEL 6: CONDITIONAL ASSOCIATION

   

γ/σ = ζ/σ = 0.5

0.906/0.854

0.119/0.037

0.127/0.045

  1. TDR is given in pairs corresponding to RF and MARS, respectively.