Skip to main content

Table 7 Independent test results (sbSVM was trained on dataset 2).

From: A semi-supervised boosting SVM for predicting hot spots at protein-protein Interfaces

Methods

Recall

Precision

Specificity

Accuracy

F1

KFC

0.33

0.42

0.79

0.65

0.37

Robetta

0.39

0.58

0.87

0.72

0.46

FOLDEF

0.26

0.48

0.87

0.69

0.33

MINERVA

0.46

0.69

0.91

0.77

0.55

KFC2

0.74

0.56

0.74

0.74

0.64

sbSVM

0.82

0.51

0.64

0.70

0.63