Table 3

Accuracy of discriminating transmembrane segments from non-transmembrane segments in trans-membrane proteins using the SOGR-I and SOGR-IB classifiers, a decision tree classifier (C4.5), and a support vector machine classifier (SVMlight version 6.01), based on ten-fold cross-validation. Three features were used, namely hydropathy (Liu-Deber scale), polarity (Grantham scale), and flexibility.

C4.5

Fold

SOGR-I

SOGR-IB

Before Pruning

After Pruning

SVM


1

72.2311

72.2311

72.4960

72.5490

72.9730

2

69.0476

67.1733

67.8318

67.6798

67.3252

3

77.1277

76.9149

77.5532

77.6596

77.7660

4

81.8913

84.5875

83.7827

83.7827

83.4608

5

79.3146

78.4889

78.3237

78.4476

78.1586

6

81.4600

83.6230

82.8119

83.1595

82.0780

7

75.9410

76.8266

75.6458

75.9410

76.3100

8

78.3488

79.2783

79.8797

79.9891

79.2783

9

64.1365

65.0418

64.5543

64.5543

64.7632

10

67.2325

65.6089

66.8635

67.0111

67.6753

Mean

74.7

75.0

75.0

75.1

75.0

Std. dev.

6.2

7.2

6.8

6.8

6.5


Reproduced with permission from [38]

Yang et al. BMC Genomics 2008 9(Suppl 1):S7   doi:10.1186/1471-2164-9-S1-S7

Open Data