The Structural Data Bank for Cross-Reactivity is a curate repository of three-dimensional structures of MHC: epitope complexes, focused on immunogenicity, similarity relationships and cross-reactivity prediction.

Cellular immunity has a major role in the surveillance against viral infections. Through the interaction between the T Cell Receptor (TCR) and the Major Histocompatibility Complex (MHC), Cytotoxic T Lymphocytes (CTL) are able to recognize viral peptides presented in the cells surface, and thus, eliminate the infected cells. Each TCR is able to interact with thousands of MHC:epitope complexes, but few of these interactions will trigger a CTL response. The capacity of a given T lymphocyte to recognize
different peptides in the cleft of a given MHC is defined as Cross-Reactivity, and it has great influence over phenomena such as heterologous immunity and molecular mimicry.

The CrossTope is a curate repository of three-dimensional structures of MHC:epitope complexes, focused on immunogenicity, similarity relashionships and cross-reactivity prediction. Our main idea is that immunogenic epitopes share characteristics that stimulates a Citotoxic T Lymphocyte to respond against
different peptides presented by a given MHC allele. The complexes hosted by these databank were obtained by large-scale in silico construction of three-dimensional models of MHC:peptide complexes, using a new approach developed by our group (D1-EM-D2, Antunes et al, 2010). This structural dataset allowed us to verify that the TCR-interaction surface of pMHC complexes has a shared pattern of charge distribution among complexes with recognized cross-reactivity.