SUMOhunt: Combining Spatial Staging between Lysine and SUMO with Random Forests to Predict SUMOylation
Figure 3
Area under the ROC (-axis: true positive rate (fraction of true positives out of total positives) and -axis: false positive rate (fraction of false positives out of total negatives)) of 10-fold cross-validation, self-consistency test, and LOOCV-ROC curve depicts the performance of a classifier by plotting true positive rate versus the false positive rate. The greater the area under the curve, higher is the performance of a classifier. For 10-fold cross-validation 89% area under the curve is obtained, for LOOCV it is 88%, while for self-consistency it is a full 100%. At the end, ROC for classification of test set has been done with an AUC of 1.
