Skip to main content

Table 3 Compared RPI-SE with other computational methods on RPI369, RPI488 and RPI1807 data sets

From: RPI-SE: a stacking ensemble learning framework for ncRNA-protein interactions prediction using sequence information

Data sets

Methods

Acc(%)

TPR(%)

TNR(%)

PPV(%)

MCC(%)

AUC

RPI369

IPMiner

75.2

73.5

79.1

71.3

50.7

0.773

RPISeq-RF

70.4

70.5

70.2

70.7

40.9

0.767

lncPro

70.4

70.8

69.6

71.3

40.9

0.740

RPI-SAN

74.9

74.1

78.7

71.7

50.4

0.778

RPI-SE

88.44

83.69

95.87

80.85

77.73

0.924

RPI488

IPMiner

89.1

93.9

83.1

94.5

78.4

0.914

RPISeq-RF

88.0

92.6

82.2

93.2

76.2

0.903

lncPro

87.0

90.0

82.7

91.0

74.0

0.901

RPI-SAN

89.7

94.3

83.7

95.2

79.3

0.920

RPI-SE

89.30

94.49

83.48

95.15

79.31

0.904

RPI1807

IPMiner

98.6

98.2

99.3

97.8

97.2

0.998

RPISeq-RF

97.3

96.8

98.4

96.0

94.6

0.996

lncPro

96.9

96.5

98.1

95.5

93.8

0.994

RPI-SAN

96.1

93.6

99.9

91.4

92.4

0.999

RPI-SE

96.86

96.71

97.69

95.83

93.65

0.994