MHC Matchmaker: An in silico based algorithm to analyze cross-species NHP, pig, and human MHC compatibility on the amino acid level.

Xenotransplantation received US Food and Drug Administration approval for clinical trials in humans, and highly sensitized patients-those with antibodies against major histocompatibility complex (MHC)-are among the groups that stand to benefit. There is no method to characterize and compare the MHC differences across different species. A database of all pig, rhesus, and human amino acid MHC sequences in the Immuno Polymorphism Database-MHC database was constructed, aligned against a common consensus, relative solvent accessibility scores calculated, and the results compared to a list of eplets from the human leukocyte antigen (HLA) Eplet Registry. Performance was compared to HLAMatchmaker using a linear regression of a random sampling of 1000 HLA alleles. Results compared favorably to HLAMatchmaker, with Pearson correlation coefficients of r2 of 0.72 and 0.74 for class I and class II MHC, respectively. We identified several HLA amino acid motifs registered in the HLA Eplet Registry that were not found across swine leukocyte antigen alleles. Our algorithm successfully compares and ranks MHC similarities and disparities across species for xenotransplantation or nonhuman primate-to-nonhuman primate allotransplantation. For highly sensitized patients, this tool may aid in risk assessments and predictions for negative crossmatch tests.

Duke Scholars

Author Annette M Jackson Surgery, Abdominal Transplant Surgery

Author Bruce W Rogers Surgery, Surgical Sciences

Author Joseph Ladowski