Table 1 The 5-fold cross-validation results were generated on the Yeast dataset by using the CNN-FSRF method.

From: Predicting Protein-Protein Interactions from Matrix-Based Protein Sequence Using Convolution Neural Network and Feature-Selective Rotation Forest

Test set

Accu.(%)

Sen.(%)

Spec. (%)

Prec.(%)

Fscore(%)

MCC(%)

AUC(%)

1

97.36

99.73

95.04

95.18

97.40

94.83

96.97

2

98.17

99.82

96.55

96.59

98.18

96.39

97.92

3

97.45

99.73

95.19

95.36

97.50

95.00

97.17

4

97.27

99.29

95.22

95.48

97.35

94.62

97.13

5

98.48

99.47

97.46

97.58

98.52

96.98

98.52

Average

97.75

99.61

95.89

96.04

97.79

95.57

97.54

Standard Deviation

0.54

0.22

1.07

1.02

0.53

1.05

0.66