Figure 5

Feature importance. Plot of percentage increase of the prediction error if the corresponding feature is randomly permuted, using random forests for regression [42]. Of all features in the sss feature set, the relative population of conformational state E (VASM830103, [38]), the estimated gas-phase basicity (GB500, [36]), and the theoretical mass lead to the highest increase of the error if the peptide's values are permuted. The number of positive charges (FAUJ880111, [41]) and the number of glutamine residues (Q) are rated the least important features.