Skip to main content

Table 1 Classification results for each PTM-specific dataset using conventional features and the SVM as a classifiera

From: A machine learning strategy for predicting localization of post-translational modification sites in protein-protein interacting regions

 

F1

Sn

Sp

PPV

ACC

AUC

MCC

Acetylation

 Hydropathy

0.51

0.51

0.51

0.51

0.51

0.50

0.01

 Secondary structure

0.49

0.49

0.50

0.49

0.49

0.50

−0.01

 Conservation

0.55

0.57

0.48

0.53

0.53

0.54

0.06

 Combined features

0.53

0.53

0.54

0.53

0.54

0.55

0.07

Phosphorylation

 Hydropathy

0.53

0.58

0.40

0.49

0.49

0.48

−0.03

 Secondary structure

0.48

0.45

0.58

0.52

0.52

0.53

0.03

 Conservation

0.53

0.55

0.45

0.50

0.50

0.51

0.01

 Combined features

0.55

0.56

0.52

0.54

0.54

0.55

0.08

Ubiquitylation

 Hydropathy

0.52

0.53

0.49

0.51

0.51

0.52

0.02

 Secondary structure

0.48

0.46

0.52

0.49

0.49

0.48

−0.02

 Conservation

0.57

0.59

0.50

0.54

0.55

0.56

0.09

 Combined features

0.54

0.54

0.53

0.53

0.53

0.55

0.06

  1. aThis table shows results when datasets were balanced (see Methods). The results using unbalanced datasets are shown in Additional file 7: Table S1