Figure 5.

Cross-validated Precision-Recall (PR) curves and Area Under the Curve (AUC) of the best-performing predictive model (RF) across different thresholds for sequence similarity. Grouped, nested fourfold cross-validation was performed to tune the hyperparameters in the inner loop and compute weighted averaged precision and recall over all classes in the outer loop. This was repeated for different thresholds of sequence similarity in the dataset that controlled the grouping in the cross-validation (i.e. the lower the threshold, the more sequences were grouped into the same fold making test set predictions more difficult). In addition to plotting the PR curves, the AUC was computed as well (see legend).