Skip to main content

Table 5 The performance of SVM based model on the real dataset1

From: MP4: a machine learning based classification tool for prediction and functional annotation of pathogenic proteins from metagenomic and genomic datasets

Accuracy (%)

79.22

Class 1

Class 2

Class 3

Sensitivity

0.88

0.74

0.77

Specificity

0.99

0.89

0.80

Balanced accuracy

0.93

0.82

0.79

  1. Where, Class 1: Non-pathogenic Proteins; Class 2: Antibiotic Resistance and Toxic proteins and Class 3: Secretory and capsular proteins