Table 2.
Predictive performance of TOP-n-gram, adapted normal distribution bi-profile Bayes (ANBPB), and parallel correlation pseudo amino acid composition (PC-PseAAC) using the jackknife test.
| Feature | Sn (%) | Sp (%) | Acc (%) | MCC |
|---|---|---|---|---|
| AAPC(320) | 81.42 | 76.44 | 78.1 | 0.55 |
| AAPC(320) + TOP-n-gram | 80.33 | 100 | 93.43 | 0.86 |
| AAPC(320) + ANBPB | 64.48 | 100 | 88.14 | 0.74 |
| AAPC(320) + PC-PseAAC | 84.15 | 72.60 | 76.46 | 0.54 |
| AAPC(320) + TOP-n-gram + ANBPB | 70.49 | 98.36 | 89.05 | 0.75 |
| AAPC(320) + TOP-n-gram + PC-PseAAC | 87.43 | 75.34 | 79.38 | 0.59 |
| AAPC(320) + ANBPB + PC-PseAAC | 75.41 | 100 | 91.79 | 0.82 |
| AAPC(320) + TOP-n-gram + ANBPB + PC-PseAAC | 94.54 | 100 | 98.18 | 0.96 |