Skip to main content

Table 2 The performance of various SVM models

From: Support Vector Machine-based method for predicting subcellular localization of mycobacterial proteins using evolutionary information and motifs

 

Cytoplasmic

Integral Membrane

Secretory

Membrane-attached

Overall accuracy

Average accuracy

Input Pattern

ACC ± sd

MCC

ACC ± sd

MCC

ACC ± sd

MCC

ACC ± sd

MCC

ACC

ACC

Amino Acid Composition

88.82 ± 5.4

0.77

86.07 ± 7.5

0.71

44.00 ± 42.2

0.57

55.00 ± 19.4

0.58

82.51

68.47

Dipeptide Composition

89.41 ± 7.8

0.72

81.09 ± 7.5

0.67

50.00 ± 36.8

0.60

50.00 ± 17.4

0.57

80.39

67.63

PSSM profile

94.71 ± 4.8

0.85

87.81 ± 6.1

0.80

44.00 ± 42.2

0.48

68.33 ± 28

0.69

86.62

73.71

  1. ACC: Accuracy; MCC: Matthews correlation coefficient; sd: Standard Deviation