Skip to main content

Table 4 Results of 10-fold cross validation with datasets of virus-host PPIs using different combinations of features

From: A generalized approach to predicting protein-protein interactions between virus and host

Features SN(%) SP(%) ACC(%) PPV(%) NPV(%) MCC AUC
RFAT 82.85 84.04 83.45 83.84 83.05 0.668 0.903
FDAT 68.34 57.84 63.11 61.86 64.65 0.264 0.689
AC 59.85 68.11 63.98 65.24 62.92 0.281 0.698
Composition 71.79 55.79 63.79 61.89 66.42 0.279 0.685
Transition 74.05 55.72 64.88 62.58 68.23 0.302 0.713
Distribution 71.79 31.55 51.67 51.19 52.80 0.036 0.515
RFAT+FDAT+AC 84.73 85.62 85.18 85.49 84.86 0.703 0.920
Composition+Transition +Distribution 76.51 61.72 69.12 66.65 72.43 0.386 0.787
All 6 features 85.36 85.92 85.64 85.84 85.44 0.712 0.925
  1. SN: sensitivity, SP: specificity, ACC: accuracy, PPV: positive predictive value, NPV: negative predictive value, MCC: Matthews correlation coefficient, AUC: area under the ROC