Skip to main content

Table 3 Our mothod performance measured by 5-fold cross validation

From: Deep learning for DNase I hypersensitive sites identification

Methods

Sn(%)

Sp(%)

ACC(%)

MCC

SVM-Revchmer [42]

82.54

79.78

81.66

0.634

PseDNC-SVM [43]

81.30

78.91

80.11

0.602

iDHS-EL [44]

81.24

76.11

78.61

0.572

Unb-PseTNC [45]

86.48

83.74

85.11

0.702

pDHS-ELM [41]

89.17

87.78

88.48

0.717

ours

88.25

96.49

92.88

0.856

  1. Note: The datasets were downloaded from [41]