Weighted-SAMGSR: combining significance analysis of microarray-gene set reduction algorithm with pathway topology-based weights to select relevant genes

Table 2 Simulation results

	Training set		Test set
Method (Size^a)	HDAC1 (%)	GNAS (%)	Error (%)	GBS	BMC	AUPR
A. Simulated from 60 independent normal-distributed random variables
SAMGSR (3.8)	19	100	16.5	0.118	0.733	0.921
W-SAMGSR (6.23)	65	100	13.2	0.101	0.755	0.948
B. Simulated based on the NSCLC microarray data
SAMGSR (3.94)	0	100	44.5	0.256	0.517	0.550
W-SAMGSR (6.28)	77	100	40.5	0.241	0.534	0.621

Note: W-SAMGSR stands for weighted-SAMGSR
^astands for average the number of genes selected by either SAMGSR or W-SAMGSR over 100 replicates

ISSN: 1745-6150