Table 3 Performance comparisons of different methods on the independent test dataset check390. The best results for each metric are in bold
From: Identifying GPCR-drug interaction based on wordbook learning from sequences
Method | Sn (%) | Sp (%) | Acc (%) | Str(%) | MCC | Threshold |
|---|---|---|---|---|---|---|
iGPCR-Drug | 80.8 | 66.9 | 71.6 | 73.9 | 0.45 | N/A |
OET-KNN | 67.7 | 84.2 | 78.7 | 76.9 | 0.52 | 0.5 |
QuickRBF | 76.2 | 77.7 | 77.2 | 77.6 | 0.52 | 0.45 |
SVM | 76.2 | 78.9 | 78.0 | 77.6 | 0.53 | 0.42 |
RF | 78.5 | 78.1 | 78.2 | 78.3 | 0.54 | 0.51 |
RF + PPP | 83.1 | 79.6 | 80.8 | 81.3 | 0.60 | 0.51 |
Proposed(Base learner) | 83.9 | 80.0 | 81.3 | 81.9 | 0.61 | 0.5 |
Proposed(Ensemble model with Ne = 4) | 83.1 | 82.7 | 82.8 | 82.9 | 0.63 | 0.5 |