Skip to main content

Table 6 Performance on balanced dataset for MTS vs SP vs N-signal-free protein prediction using SVM classifier

From: Plus ça change – evolutionary sequence divergence predicts protein subcellular localization signals

 

Divergence

Classical features

Combination

 

AUC

MCC

AUC

MCC

AUC

MCC

MTS

0.67±0.10

0.35±0.20

0.84±0.07

0.68±0.13

0 . 8 8±0.05

0 . 7 8±0.09

SP

0.71±0.09

0.41±0.16

0.92±0.05

0.85±0.10

0 . 9 4±0.01

0 . 8 8±0.03

N-signal-free

0.79±0.07

0.60±0.13

0.78±0.09

0.57±0.18

0 . 8 6±0.07

0 . 7 4±0.13

% accuracy

62.86±5.84

79.92±5.54

8 6 . 1 9±4.67

  1. The 5-fold cross-validation performance of an SVM classifier using: divergence features only, classical features only, and the two combined; is shown for three-way classification on a balanced dataset (53 proteins from each class, yeast curated orthologs).