Table A1.
Comparison of prediction accuracy of in the CASP10 test dataset.
| Sen | Spec | Acc | Acc-b | MCC | F1 | AUC_ROC | AUC_PR | |
|---|---|---|---|---|---|---|---|---|
| Disembl | 0.379 | 0.954 | 0.929 | 0.666 | 0.286 | 0.318 | 0.754 | 0.295 |
| IUPred | 0.168 | 0.958 | 0.924 | 0.563 | 0.122 | 0.161 | 0.618 | 0.175 |
| VSL2 | 0.612 | 0.811 | 0.803 | 0.712 | 0.214 | 0.213 | 0.774 | 0.275 |
| Espritz | 0.512 | 0.921 | 0.903 | 0.716 | 0.298 | 0.316 | 0.815 | 0.404 |
| DISOPRED3 | 0.362 | 0.993 | 0.966 | 0.678 | 0.495 | 0.481 | 0.860 | 0.495 |
| PONDRFIT | 0.586 | 0.929 | 0.914 | 0.758 | 0.362 | 0.374 | 0.830 | 0.358 |
| MFDp2 | 0.325 | 0.975 | 0.947 | 0.650 | 0.322 | 0.349 | 0.778 | 0.352 |
| IUPred2 | 0.164 | 0.959 | 0.924 | 0.561 | 0.119 | 0.158 | 0.616 | 0.170 |
| AUCpreD | 0.425 | 0.984 | 0.960 | 0.705 | 0.465 | 0.481 | 0.863 | 0.501 |
| This work | 0.629 | 0.840 | 0.831 | 0.734 | 0.249 | 0.245 | 0.793 | 0.305 |
Note Bene. DISOPRED3 is one of the two best predictors in CASP10 competition and therefore was used in the comparison. The other one and its web server were not available when the study was carried out. The highest value in each of these measures is in bold and highlighted (red).