Table 6.

Simulation results for selecting k important variables among 2,000 candidates including the most and least important surrogate variables across 600 subjects.

	Bonferroni			FDR			TT
	n.sv = 5	n.sv = 10	n.sv = 15	n.sv = 5	n.sv = 10	n.sv = 15	n.sv = 5	n.sv = 10	n.sv = 15
Most important surrogate variables included
	# incorrect
k = 10	0	0	0	6	6	6	5	4	3
k = 100	2	3	2	19	19	19	7	5	5
k = 200	6	6	5	26	29	29	6	6	4
k = 400	12	11	11	40	38	39	7	7	7
	Sensitivity
k = 10	1	1	1	1	1	1	1	1	1
k = 100	0.98	0.97	0.98	1	1	1	0.98	0.98	0.98
k = 200	0.97	0.97	0.975	0.995	0.995	0.995	0.985	0.985	0.985
k = 400	0.97	0.973	0.973	1	1	1	0.988	0.988	0.985
	Specificity
k = 10	1	1	1	0.997	0.997	0.997	0.997	0.998	0.998
k = 100	1	1	1	0.99	0.99	0.99	0.997	0.998	0.998
k = 200	1	1	1	0.986	0.984	0.984	0.998	0.998	0.999
k = 400	1	1	1	0.975	0.976	0.976	0.999	0.999	0.999
Most important surrogate variables not included
	# incorrect
k = 10	10	10	10	10	10	10	10	10	10
k = 100	100	100	100	100	100	100	100	100	100
k = 200	200	200	200	200	200	200	200	200	200
k = 400	400	400	400	400	400	400	400	400	400
	Sensitivity
k = 10	0	0	0	0	0	0	0	0	0
k = 100	0	0	0	0	0	0	0	0	0
k = 200	0	0	0	0	0	0	0	0	0
k = 400	0	0	0	0	0	0	0	0	0
	Specificity
k = 10	1	1	1	1	1	1	1	1	1
k = 100	1	1	1	1	1	1	1	1	1
k = 200	1	1	1	1	1	1	1	1	1
k = 400	1	1	1	1	1	1	1	1	1

FDR: false discovery rate, TT: training and testing, n.sv = number of surrogate variables, and k: the number of truly important CpG sites out of 2,000.