Fig. 4

Analysis of classification errors. A Histograms of 672 mutations in folding energy dataset and 515 mutations in binding energy dataset show the unequal distributions of stabilizing (\({\mathrm{\Delta \Delta }G}_{exp}<\) − 0.5 kcal/mol), neutral (between − 0.5 and 0.5 kcal/mol), and destabilizing (> 0.5 kcal/mol). B The relative frequencies of the stabilizing, neutral, and destabilizing mutations are plotted along intervals of \({\mathrm{\Delta \Delta }G}_{FoldX}\). Shaded areas indicate incorrect prediction by FoldX. FoldX tends to predict destabilizing mutations correctly than stabilizing mutations. C Same sets of mutations are classified according to the interval \({\mathrm{\Delta \Delta }G}_{FoldX}\) ± error bounds, and their relative frequencies are expressed along \({\mathrm{\Delta \Delta }G}_{FoldX}\)