Skip to main content
. 2021 Feb 15;223(Suppl 1):S32–S37. doi: 10.1093/infdis/jiaa333

Figure 1.

Figure 1.

Application of structure-based network analysis to HIV protein structures to delineate mutationally constrained CD8+ T-cell epitopes. Structure-based network analysis utilizes atomic-level protein crystal structure data, in this case for HIV reverse transcriptase (PDB: 3KJV), to construct a protein network by calculating and summing all noncovalent interactions between residue side chains. The width of the red edges indicates interaction strength. The protein network is then used to calculate network scores for each amino acid residue, indicating its topological importance to the overall structure of the protein (node size indicates relative network score). Using these scores, linear CD8+ T-cell epitopes can then be quantified on the basis of their topological importance as either highly networked or poorly networked. Recent work has demonstrated that CTL targeting of highly networked epitopes is associated with immune control, while targeting of poorly networked epitopes is associated with disease progression [44].