A Bayesian Hierarchical Model for Relating Multiple SNPs within Multiple Genes to Disease Risk

Table 1

Simulation analysis based on 10 parameter replicates with 10 data replicates per parameter replicate.

(a)  

SNPTrue	Average counts^a	Posterior SNP inclusion^b			BF^c
		−1	0	1	>3	>20	>150
−1	17.5	24.14%	71.75%	4.11%	25.54%	17.49%	12.46%
0	348.1	3.19%	93.76%	3.05%	3.90%	0.68%	0.19%
1	18.4	3.88%	70.19%	25.94%	28.15%	19.13%	15.54%

(b) 

GeneTrue	Average counts^d	Posterior gene inclusion^e		BF^f
		Not included	Included	>3	>20	>150
Not included	13.9	55.95%	44.05%	3.67%	0.58%	0.15%
Included	24.1	36.55%	63.45%	27.71%	20.01%	17.14%

aAverage counts of simulated SNP inclusion indicators based on 10×10 replicates.
bAverage row percentages of the distribution of posterior SNP inclusion indicators based on 10×10 replicates.
cAverage row percentages of the SNP counts among the range of the indicated Bayes factors based on 10×10 replicates.
dAverage counts of simulated gene inclusion indicators based on 10×10 replicates.
eAverage row percentages of the distribution of posterior gene inclusions based on 10×10 replicates.
fAverage row percentages of the gene counts among the range of the indicated Bayes factors based on 10×10 replicates.