Skip to main content

Table 1 The performance of the SVMs that were trained with the top hundred genes as ranked by the Netrank on the development set and tested with the test set

From: Netrank: network-based approach for biomarker discovery

Cancer type

Precision

Recall

F1_score

Accuracy

AUC

No. samples

Development set

Test set

Case

Control

Case

Control

ACC

0.99

0.95

0.97

0.99

0.95

57

2314

21

996

BLCA

0.97

0.79

0.85

0.98

0.79

95

2276

41

976

BRCA

0.97

0.97

0.97

0.98

0.97

618

1753

244

773

CESC

0.94

0.95

0.94

0.99

0.97

91

2280

33

984

CHOL

0.99

0.81

0.88

0.99

0.82

25

2346

11

1006

KICH

0.99

0.92

0.96

0.99

0.93

44

2327

21

996

KIRP

0.99

0.98

0.99

0.99

0.99

88

2283

43

974

LGG

1

0.95

0.97

0.99

0.95

16

2355

11

1006

LIHC

0.99

0.95

0.97

0.99

0.96

132

2239

67

950

LUAD

0.99

0.95

0.97

0.99

0.95

213

2158

86

931

MESO

0.97

0.89

0.93

0.99

0.9

54

2317

25

992

PAAD

0.92

0.95

0.93

0.99

0.96

120

2251

36

981

PRAD

0.99

0.99

0.99

0.99

0.99

141

2230

82

935

SARC

0.98

0.94

0.96

0.99

0.94

166

2205

71

946

SKCM

0.97

0.97

0.97

0.99

0.98

250

2121

99

918

TGCT

0.98

0.98

0.98

0.99

0.99

95

2276

38

979

THYM

0.99

0.97

0.98

0.99

0.97

83

2288

34

983

UCS

0.99

0.71

0.79

0.98

0.71

38

2333

19

998

UVM

0.97

0.99

0.98

0.99

1

61

2310

19

998

  1. The numbers of positive and negative samples in the training and testing sets are reported