|
Resolution: standard / high Figure 4.
ROC curves for the sequence-based predictor. ROC curves for the sequence-based predictor. The cross-validation True Positive Rate (TPR) versus the False Positive Rate (FPR) are plotted the best method (SVM-WIN31) and for the baseline method (SVM-BASE). In part (A) the ROC curves of the two methods are relative to the prediction of increasing and decreasing free energy mutations (|ΔΔG|>1.0 kcal/mole), while in part (B) they are calculated for neutral mutations (|ΔΔG|≤1.0 kcal/mole).
Capriotti et al. BMC Bioinformatics 2008 9(Suppl 2):S6 doi:10.1186/1471-2105-9-S2-S6 |