Figure 5
From: Functional site prediction selects correct protein models
Specificity-Sensitivity plots for five training-set proteins. The Specificity-Sensitivity curves for each of the five training-set proteins (a) 3chy, (b) 1coz, (c) 2trx, (d) 1f4p and (e) 1di0 using "Fold Score" (red) are shown. Sensitivity = TP/(TP+FN) and Specificity = TN/(TN+FP). i.e. Specificity is defined as the fraction of significant hits (hits with scores above a threshold) being correct. Sensitivity is defined as the fraction of possible correct hits being significant. (TP = True Positives, TN = True Negatives, FP = False Positives, FN = False Negatives).