Table 2.
Performance parameters of the established prediction models for spike protein, nucleocapsid protein, and 2′-O-ribose-methyltransferase.
| TP | TN | FP | FN | Sensitivity | Specificity | Overall predictive accuracy | Precision | AUC | |
|---|---|---|---|---|---|---|---|---|---|
| Learning | |||||||||
| Spike protein (neural network) | 16 | 19 | 1 | 0 | 1.000 | 0.950 | 0.972 | 0.941 | 0.994 |
| Nucleocapsid protein (neural network) | 15 | 19 | 1 | 1 | 0.938 | 0.950 | 0.944 | 0.938 | 0.997 |
| 2-o-ribose-methyltransferase (naïve bayes) |
16 |
18 |
2 |
2 |
0.889 |
0.900 |
0.895 |
0.889 |
0.978 |
| External validation | |||||||||
| Spike protein | 8 | 10 | 0 | 0 | 1.000 | 1.000 | 1.000 | 1.000 | |
| Nucleocapsid protein | 8 | 10 | 0 | 0 | 1.000 | 1.000 | 1.000 | 1.000 | |
| 2-o-ribose-methyltransferase | 9 | 10 | 0 | 0 | 1.000 | 1.000 | 1.000 | 1.000 |
TP, true positive; TN, true negative; FP, false positive; FN, false negative; AUC, area under the curve.