Genotype-Based Bayesian Analysis of Gene-Environment Interactions with Multiple Genetic Markers and Misclassification in Environmental Factors

Table 2

Biases and root mean squared errors (RMSEs) of risk parameters obtained based on pseudo-MLE and the proposed MCMC. The results are based on 500 samples of 350 cases and 350 controls. Genotype is simulated at the two marker loci with 𝑃_𝑀𝑖=0.25, 𝑖=1,2. The environmental covariate (𝑋) is binary and measured with error misclassification probabilities being 0.20 for exposed and 0.25 for nonexposed subjects. The data is simulated and analyzed under the additive effect model and the LD measure Δ_𝑀1𝑀2=0.03.

Parameter	True value	Pseudo-MLE		MCMC
		Bias	RMSE	Bias	RMSE
β X	1.099	0.035	0.392	0.013	0.236
β A1	0.406	−0.268	1.035	−0.079	0.397
β A2	0.789	−0.319	1.062	−0.085	0.372
β A3	0.693	−0.293	1.043	−0.092	0.365
β AX1	0.916	0.432	1.135	0.103	0.432
β AX2	0.693	0.391	1.047	0.085	0.481
β AX3	1.099	0.293	1.113	0.097	0.427