Table 5.
PaPI performances on the unpredictable variants by PolyPhen2 and SIFT
| Test Set | Tool | AUC | Accuracy [IC95%] | Sens | Spec | PPV | NPV | F-m | MCC |
|---|---|---|---|---|---|---|---|---|---|
| # 1 | PaPI (RF) | .9368 | .8676 [.8420-.8896] | .9171 | .8245 | .8198 | .9196 | .8657 | .7405 |
| # 2 | PaPI (RF) | .9418 | .8611 [.8352-8836] | .9214 | .8077 | .8095 | .9205 | .8619 | .7296 |
| # 3 | PaPI (RF) | .942 | .8830 [.8523-9080] | .9256 | .845 | .8421 | .9271 | .8819 | .7699 |
PaPI performances on the three test retaining only those variants unpredictable both for PolyPhen2 and SIFT. In this case, PaPI coincides with RF. Area under the curve (AUC), accuracy with 95% confidence intervals, sensitivity (Sens), specificity (Spec), Positive Predictive Value (PPV), Negative Predictive Value (NPV), F-measure (F-m) and Matthews correlation coefficient (MCC) are reported for each method.