Skip to main content

Table 1 Localization predictions of a set of 299 experimentally verified proteins with known localization

From: Genome-wide subcellular localization of putative outer membrane and extracellular proteins in Leptospira interrogans serovar Lai genome using bioinformatics approaches

Actual localization

Total

TP

FP

FN

TN

Precision

Recall

PSORTb

       

CP

145

110

1

35

110

99.10%

75.86%

CM

69

55

2

14

166

96.49%

79.71%

PP

29

18

0

11

207

100.00%

62.07%

OM

38

30

0

8

195

100.00%

78.95%

EX

18

6

3

12

216

66.67%

33.33%

Total

299

219

6

80

894

97.33%

73.24%

Proteome Analyst

       

CP

145

94

0

51

119

100.00%

64.83%

CM

69

59

2

10

162

96.72%

85.51%

PP

29

19

3

10

201

86.36%

65.52%

OM

38

31

0

7

192

100.00%

81.58%

EX

18

9

6

9

207

60.00%

50.00%

Total

299

212

11

87

881

95.07%

70.90%

ProtCompB

       

CP

145

127

11

18

144

92.03%

87.59%

CM

69

55

7

14

227

88.71%

79.71%

PP

29

19

9

10

261

67.86%

65.52%

OM

38

23

18

15

243

56.10%

60.53%

EX

18

11

4

7

277

73.33%

61.11%

Total

299

235

49

64

1152

82.75%

78.60%

Consensus vote

       

CP

145

67

0

78

154

100.00%

46.21%

CM

69

43

0

26

230

100.00%

62.32%

PP

29

11

0

18

270

100.00%

37.93%

OM

38

19

0

19

261

100.00%

50.00%

EX

18

4

0

13

216

100.00%

23.53%

Total

299

144

0

154

1131

100.00%

48.32%

Majority vote (2 out of 3 predictions)

       

CP

145

121

0

24

154

100.00%

83.45%

CM

69

59

0

10

230

100.00%

85.51%

PP

29

17

0

12

270

100.00%

58.62%

OM

38

29

0

10

213

100.00%

74.36%

EX

18

6

1

12

215

85.71%

33.33%

Total

299

232

1

68

1082

99.57%

77.33%

Combination method

       

CP

N/A

N/A

N/A

N/A

N/A

N/A

N/A

CM

N/A

N/A

N/A

N/A

N/A

N/A

N/A

PP

29

25

0

4

56

100.00%

86.20%

OM

38

34

1

4

46

97.14%

89.47%

EX

18

12

2

6

65

85.71%

66.67%

Total

85

71

3

14

167

95.95%

87.53%

  1. 299 proteins obtained from the test set used in comparison study by Gardy and Brinkman [22] Majority vote is the result from 2 out of 3 predictions. Combination method: the result from non-consensus vote with significant DBsubloc [31] and/or GTD prediction [32] Precision is calculated as TP/(TP+FP), Recall is calculated as TP/(TP+FN) TP = true positive, TN = true negative, FP = false positive, FN = false negative, N/A= Not applicable