Skip to main content

Table 6 Comparison of the prediction performance between our proposed method and other state-of-the-art works on S.cerevisiae dataset

From: Predicting protein-protein interactions via multivariate mutual information of protein sequences

Method Feature Classifier ACC(%) SN(%) PPV(%) MCC(%)
Our method MMI+NMBAC RF 95.01 ±0.46 92.67 ±0.50 97.16 ±0.55 90.10 ±0.92
You’s work [18] MLD RF 94.72 ±0.43 94.34 ±0.49 98.91 ±0.33 85.99 ±0.89
You’s work [30] AC+CT+LD+MAC E-ELM 87.00 ±0.29 86.15 ±0.43 87.59 ±0.32 77.36 ±0.44
You’s work [16] MCD SVM 91.36 ±0.36 90.67 ±0.69 91.94 ±0.62 84.21 ±0.59
Wong’s work [17] PR-LPQ Rotation Forest 93.92 ±0.36 91.10 ±0.31 96.45 ±0.45 88.56 ±0.63
Guo’s work [12] ACC SVM 89.33 ±2.67 89.93 ±3.68 88.87 ±6.16 N/A a
Guo’s work [12] AC SVM 87.36 ±1.38 87.30 ±4.68 87.82 ±4.33 N/A a
Zhou’s work [14] LD SVM 88.56 ±0.33 87.37 ±0.22 89.50 ±0.60 77.15 ±0.68
Yang’s work [15] LD KNN 86.15 ±1.17 81.03 ±1.74 90.24 ±1.34 N/A a
  1. aN/A means not available