It would be nice to have a similar CDR-fragment-style annotation of relevant MHC structural components. This would include residues that make up the TCR binding portion of the MHC and residues that make up the peptide binding portion of the MHC molecule. These would be determined and set based on the wide-scale analysis of existing crystal structures.