Large-scale risk prediction applied to Genetic Analysis Workshop 17 mini-exome sequence data

BMC Proceedings

Table 1 Prediction rule of three proposed methods

Feature	Empirical Bayes method			Weighted empirical Bayes method				Joint covariance model
	Genes	#SNP	MAF	Genes	#Syn SNP	#Non SNP	MAF	Genes	#Syn SNP	#Non SNP	MAF
1	Age			Age				Age
2	Smoke			Smoke				Smoke
3	ATP11A	1	0.29	SUSD2	13	23	<0.01	ATP11A	1		0.29
					2	4	0.01–0.05
					1	2	≥0.05
4	*FLT1*	25	<0.01	*FLT1*	8	17	<0.01	BUD13	1		0.11
		7	0.01–0.05		5	2	0.01–0.05
		3	≥0.05		2	1	≥0.05
5	SUSD2	36		ATP11A	1		0.29	*C10ORF107*	1		0.13
		6
		3
6	BUD13	1	0.11	RIPK3	4	13	<0.01	RIPK3	4	13	<0.01
					1	1	0.01–0.05		1	1	0.01–0.05
					1	1	≥0.05		1	1	≥0.05
7	RIPK3	17	<0.01	BUD13	1		0.11	SUSD2	13	23	<0.01
		2	0.01–0.05						2	4	0.01–0.05
		2	≥0.05						1	2	≥0.05
8	*C10ORF107*	1	0.13	ADAMTS4	10	23	<0.01	*FLT1*	8	17	<0.01
					2	2	0.01–0.05		5	2	0.01–0.05
					1	2	≥0.05		2	1	≥0.05
9	ADAMTS4	33	<0.01	WNT16	8	7	< 0.01	GPR158		1	0.1
		4	0.01–0.05		1	2	0.01–0.05
		3	≥0.05			2	≥0.05
10	MAP3K12	14	<0.01	GOLGA1	1		<0.01	ANAPC5	14	12	<0.01
		3	0.01–0.05			1	0.01–0.05		1		0.01–0.05
						1	≥0.05				≥0.05

Top 10 important features from the model incorporating genes and environmental variables for the three proposed methods. #SNP, number of SNPs within a specific gene; #Syn SNP, number of synonymous SNPs; #Non SNP, number of nonsynonymous SNPs. MAF shows three intervals of minor allele frequency: MAF < 0.01, 0.01 ≤ MAF < 0.05, and MAF ≥ 0.05. The boldfaced genes and environmental variables are real causal features that are selected across the three proposed models.

ISSN: 1753-6561