Neural networks for modeling gene-gene interactions in association studies

BMC Genomic Data

Table 3 Additive model (ADD).

	Low risk	High risk
	a₁ = 2 a₀; a₂ = 4 a₀	a₁ = 5 a₀; a₂ = 10 a₀
	b₁ = 5·b₀; b₂ = 10 b₀	b₁ = 5 b₀; b₂ = 10 b₀
Theoretical penetrance matrix
Neural network
Mean absolute difference E
Sum	0.2313	0.2059
Logistic regression
Mean absolute difference E
Sum	0.2530	0.2544
Logistic regression (design variables)
Mean absolute difference E
Sum	0.2897	0.2804

Mean absolute differences between theoretical and estimated penetrance matrices from 100 replications in the low and high risk scenario.

ISSN: 2730-6844