Skip to main content

Table 5 Results of β -turns prediction on the BT547 and BT823 dataset

From: LVQ-SMOTE – Learning Vector Quantization based Synthetic Minority Over–sampling Technique for biomedical data

 

DEBT [[18]]

DEBT + our method

Dataset

β -turn type

MCC

Sensitivity

Specificity

MCC

Sensitivity

Specificity

BT547

I

0.38

71.6%

82.6%

0.40

73.7%

85.0%

 

II

0.33

63.0%

90.8%

0.31

66.7%

86.1%

 

IV

0.27

69.8%

73.3%

0.38

81.6%

75.2%

 

VIII

0.14

47.8%

84.4%

0.26

60.3%

84.1%

 

Non-turn

0.37

21.1%

99.7%

0.39

30.4%

97.6%

BT823

I

0.39

70.6%

84.2%

0.37

71.3%

82.5%

 

II

0.33

62.7%

91.2%

0.30

61.4%

92.1%

 

IV

0.27

68.3%

74.4%

0.35

78.4%

78.9%

 

VIII

0.14

42.2%

87.2%

0.17

47.9%

86.3%

 

Non-turn

0.38

23.6%

99.7%

0.40

27.5%

97.4%