Research Article
Ensemble Learning Prediction of Drug-Target Interactions Using GIST Descriptor Extracted from PSSM-Based Evolutionary Information
Table 3
5-fold cross-validation results were generated through the proposed method on the ion channel dataset.
| Testing set | Accuracy (%) | Precision (%) | Sensitivity (%) | MCC (%) | AUC | AUPR |
| 1 | 85.59 | 86.76 | 84.12 | 75.33 | 0.9292 | 0.8435 | 2 | 86.95 | 87.63 | 86.15 | 77.30 | 0.9431 | 0.8686 | 3 | 87.63 | 88.18 | 87.29 | 78.31 | 0.9323 | 0.8633 | 4 | 85.59 | 83.16 | 87.59 | 75.31 | 0.9370 | 0.8056 | 5 | 83.90 | 86.01 | 81.73 | 72.96 | 0.9146 | 0.8287 | Average | 85.93 ± 1.44 | 86.35 ± 1.96 | 85.38 ± 2.45 | 75.84 ± 2.07 | 0.9312 ± 0.0107 | 0.8419 ± 0.0258 |
|
|