Random Subspace Aggregation for Cancer Prediction with Gene Expression Profiles

Table 1

Dataset.

Data	Feature	Sample	Class
Breast Cancer	24481	97; 78 training (34 relapse + 44 nonrelapse); 19 test (12 relapse + 7 nonrelapse)	Relapse; Nonrelapse
Leukemia	7129	72; 38 training (27 ALL + 11 AML); 34 test (20 ALL + 14 AML)	All; AML
Lung Cancer	12533	181; 32 training (16 mesothelioma + 16 ADCA); 149 test (15 mesothelioma + 134 ADCA)	Mesothelioma; ADCA
Prostate	12600	136; 102 training (52 tumor + 50 normal); 34 test (25 tumor + 9 normal)	Tumor; Normal
Colon Tumor	2000	62; 22 positive + 40 negative	Positive; Negative
CNS	7129	60; 21 Class 1 + 39 Class 0	Class 1; Class 0
Ovarian	15154	253; 162 cancer + 91 normal	Cancer; Normal
DLBCL	4026	47; 24 germinal + 23 activated	Germinal; Activated