Fig. 1
From: Performance of Web tools for predicting changes in protein stability caused by mutations
ROC and PRC for predictions made on the full dataset of monomeric proteins. Panels a, b show, respectively, the ROC and PRC for predictions made taking into account only those mutations with a ΔΔG value outside the range of the experimental error (number of positive elements is 102, number of negative elements is 443). Panels c, d show, respectively, the ROC and PRC for predictions made taking into account only those mutations with a ΔΔG value inside the range of the experimental error (number of positive elements is 79, of the negative elements is 118). TNR: True Negative Rate, TPR: True Positive Rate, PPV: Positive Predictive Value