Table 3. Per-residue performance of disorder prediction on CASP and CAMEO targets.
Methods | CASP | CAMEO | ||||||||
---|---|---|---|---|---|---|---|---|---|---|
bAcc | Sens | Spec | Mcc | AUC | bAcc | Sens | Spec | Mcc | AUC | |
IUpreda | 0.66 | 0.36 | 0.95 | 0.33 | 0.68 | 0.69 | 0.46 | 0.93 | 0.37 | 0.79 |
DisoPred3p | 0.67 | 0.36 | 0.98 | 0.47 | 0.84 | 0.70 | 0.46 | 0.95 | 0.42 | 0.83 |
DisoPred3T | 0.68 | 0.39 | 0.99 | 0.54 | 0.85 | 0.70 | 0.43 | 0.96 | 0.44 | 0.83 |
RaptorX-Propertya | 0.72 | 0.47 | 0.97 | 0.51 | 0.86 | 0.72 | 0.48 | 0.95 | 0.45 | 0.84 |
RaptorX-Propertyp | 0.76 | 0.53 | 0.98 | 0.55 | 0.89 | 0.75 | 0.53 | 0.97 | 0.49 | 0.88 |
The evaluation criteria are balanced accuracy (bAcc), sensitivity (Sens), specificity (Spec), the Matthews correlation coefficient (Mcc) and area under the ROC curve (AUC).