Skip to main content

Advertisement

Table 3 Differences between theoretic and estimated penetrance functions (models representing a masking effect of the genetic factor)

From: Artificial neural networks modeling gene-environment interaction

    High risk scenario    Low risk scenario  
   Neural network Logistic regression Logistic regression (DV) Neural network Logistic regression Logistic regression (DV)
    n = 1000 + 1000    n = 1000 + 1000  
  Model 1 38.63 211.62 105.83 41.07 195.15 87.57
g u E g u Model 2 117.94 359.10 155.40 101.92 323.89 114.71
Model 3 40.67 253.01 85.51 43.15 258.19 65.87
  Model 4 103.37 228.10 85.16 103.63 227.50 59.74
    n = 500 + 500    n = 500 + 500  
  Model 1 54.58 219.39 136.26 70.40 207.97 140.74
g u E g u Model 2 144.35 363.36 176.74 183.28 327.58 143.06
Model 3 60.98 261.86 110.93 66.25 278.61 114.68
  Model 4 143.62 235.44 102.13 115.59 237.14 81.13
    n = 200 + 200    n = 200 + 200  
  Model 1 126.56 252.88 251.70 192.47 244.17 225.63
g u E g u Model 2 262.92 371.69 230.25 297.68 348.46 215.70
Model 3 139.27 324.55 215.12 141.28 328.64 191.61
  Model 4 189.69 287.39 169.86 164.13 280.21 149.95
  1. Sum of mean absolute differences between theoretic and estimated penetrance function for 100 case-control data sets in the low and high risk scenario for different sample sizes. Bold numbers mark the best model fit comparing neural networks and logistic regression models. DV = design variables.