Microarray-based cancer prediction using single genes

BMC Bioinformatics

Table 1 The LOOCV classification accuracy (%)

Method Dataset	SGC-t	SGC-W	DLDA	k-NN	SVM	RF
Melanoma	97*	96**	97*	97*	97*	97*
Breast Cancer 1	63**	69*	61	53	52	43
Brain Cancer	80*	77**	65	73	60	70
Breast Cancer 2	58	50	73*	67**	73*	67**
Gastric Tumor	89	80	81	96**	97*	95
Lung Cancer 1	98*	95**	95**	98*	98*	98*
Lung Cancer 2	93**	93**	99*	99*	99*	99*
Lymphoma	74*	71**	66	52	59	57
Myeloma	68	67	75	78**	74	79*
Pancreatic Cancer	69**	90*	63	61	65	55
Prostate Cancer	89**	89**	78	93*	93*	93*

Note:
1 SGC-t: Single Gene Classifier with the t-test gene selection method.
2 SGC-W: Single Gene Classifier with the WMW gene selection method.
3 for each dataset, the highest classification accuracy is highlighted with a single asterisk and the second highest is highlighted with a double asterisk.

ISSN: 1471-2105