Table 1.
Five-fold cross-validation results shown using our proposed method on the Yeast dataset.
| Model | Testing Set | Accuracy | Sensitivity | Specificity | PPV | NPV | MCC |
|---|---|---|---|---|---|---|---|
| PSSM+ LR+RVM | 1 | 94.7% | 95.4% | 94.0% | 93.9% | 95.46% | 89.3% |
| 2 | 95.3% | 96.1% | 94.5% | 94.7% | 95.96% | 91.1% | |
| 3 | 93.9% | 93.9% | 93.8% | 93.8% | 93.91% | 88.5% | |
| 4 | 93.8% | 93.6% | 94.1% | 94.4% | 93.22% | 88.4% | |
| 5 | 95.1% | 94.9% | 95.2% | 94.9% | 95.2% | 90.6% | |
| Average | 94.6 ± 0.6% | 94.8 ± 1.0% | 94.3 ± 0.5% | 94.3 ± 0.4% | 94.75 ± 1.1% | 89.6 ± 1.2% | |
| PSSM+ LR+SVM | 1 | 88.3% | 87.3% | 89.3% | 88.8% | 87.8% | 79.4% |
| 2 | 89.3% | 89.4% | 89.1% | 89.2% | 89.3% | 80.8% | |
| 3 | 89.8% | 89.2% | 90.3% | 90.7% | 88.8% | 81.6% | |
| 4 | 89.7% | 88.3% | 91.2% | 90.9% | 88.6% | 81.6% | |
| 5 | 90.0% | 88.4% | 91.5% | 90.8% | 89.2% | 81.9% | |
| Average | 89.4 ± 0.6% | 88.5 ± 0.8% | 90.3 ± 1.0% | 90.1 ± 1.0% | 88.7 ± 0.5% | 81.1 ± 1.0% |
SVM: support vector machine; PSSM: position specific scoring matrix; AB: average blocks; RVM: relevance vector machine; PPV: Positive Predictive Value; NPV: Negative Predictive Value; MCC: Matthews Correlation Coefficient.