Skip to main content

Table 4 Performance comparison among the AMPs prediction methods reported in [12] with our proposed approach for the APD3 dataset

From: Optimal selection of molecular descriptors for antimicrobial peptides classification: an evolutionary feature weighting approach

Tool

Task

Sens(%)

Spec(%)

Prec(%)

Bal Acc(%)

MOEA-FW(SVM-L)

Antimicrobial

89.24

82.87

5 1 . 9 8

8 6 . 0 5

CAMPR3(RF)

 

9 4 . 8 0 a

72.65

40.30

82.49

CAMPR3(SVM)

 

90.60

72.10

39.25

81.11

ADAM

 

91.07

68.88

35.09

76.49

MLAMP

 

75.59

82.27

41.78

72.94

DBAASP

 

62.81

92.87

38.28

57.49

AMPA

 

39.17

8 4 . 7 9

39.09

66.80

MOEA-FW(SVM-L)

Antibacterial

8 1 . 9 4

9 1 . 5 5

6 5 . 9 2

8 6 . 7 5

AntiBP2

 

66.59

26.00

15.25

46.30

MOEA-FW(SVM-L)

Bacteriocin

9 3 . 1 0

92.95

71.05

93.03

BAGEL3

 

86.36

1 0 0 . 0

1 0 0 . 0

9 3 . 1 8

BACTIBASE

 

38.36

1 0 0 . 0

1 0 0 . 0

69.48

  1. aBold font indicates the best value per measure