Skip to main content

Table 4 Classification results for the score matrices with SLiMs obtained from SM using FS

From: The predictive performance of short-linear motif features in the prediction of calmodulin-binding proteins

Dataset for classification # features Classifier Accuracy (%) MCC ROC Area
S score matrix   SVM-Polynomial (c=1,g=0) 66.1 0.33 0.66
  7 RF 77.8 0.56 0.83
   k-NN (k=1) 77.0 0.54 0.77
T score matrix   SVM-Polynomial (c=1,g=0) 53.0 0.09 0.53
  9 RF 69.3 0.38 0.75
   k-NN (k=1) 66.4 0.33 0.66