Fig. 4
From: Staged heterogeneity learning to identify conformational B-cell epitopes from antigen sequences
Prediction on a glucose-dependent insulinotropic polypeptide receptor (PDB ID: 2QKH, chain: A). (a) SePre, (b) Straightforward model, (c) CBTOPE, (d) DiscoTope 2.0, (e) ElliPro, (f) SEPPA 2.0. The epitope (IEDB epitope ID: 194683) is shown as spheres: the correctly predicted epitope residues (true positive) are colored in red, while the unrecognized epitope residues (false negative) are colored in orange. The non-epitope residues predicted as epitope (false positive) are shown as yellow sticks