? At the core of transplant rejection, immune defense against viruses and autoimmunity and is the recognition of peptides in the context of MHC-I molecules by CD8 T cells and NK cells. Here, we will combine our skills in immunology, computer modeling, database design and mathematical analysis to generate libraries of predicted self and viral peptides for most major MHC-I alleles in mouse and men. We collaborate with two experimental groups to validate the predicted peptides and study their correlation with autoimmunity. The main elements we will develop and merge are estimates of peptide MHC-I binding, TAP binding and peptide cleavage probability. We will combine the resulting scores and apply them to existing genomic data. ? ? To produce the human libraries, we will develop new bioinformatics tools and merge them with existing tools. The result will be a score determining the probability of a peptide to be presented. We will apply this score to the human, mouse and to some viral genomes, in order to determine the self and non-self repertoire. The expected libraries will be validated through a collaboration with an experimental group. Each validated library will contain a list of peptides (from human, murine or viral source), which can be presented on a given MHC-I allele, accompanied by a score representing the """"""""quality"""""""" of the peptide. This score will include the estimated binding affinity to the MHC binding groove, the TAP binding probability, the cleavage probability (either by the regular or the immuno-proteasome) and the average expression level of the gene carrying the peptide. Viral peptides will also include a score representing their maximal similarity to self peptides. The resulting validated libraries will be publicly accessible through a web based server. ? ? The accurate prediction of allele specific MHC-peptide interactions may enable us to manipulate the autoimmune response in organ specific autoimmune diseases, and the immune response against viruses. It may also help immunologist improve their ability to optimize MHC-matching for transplantation. The main applications of the libraries that will be developed in the current research program are: A) A peptide vaccine development methodology. We will use the viral peptide libraries in order to detect peptides from conserved viral genomic regions present in these libraries. We will use these peptides to produce genes sequences that code for proteins having the same immunological profile as the viruses, without its pathogenic effects. These peptides are good candidates for peptide vaccines in cases where an attenuated virus is too dangerous or unavailable. B) Analysis of the profiles presented to the immune system in autoimmune diseases (mainly IDDM and MS) and the link between MHC usage and autoimmunity. This part of the research will be done in colaboration with an experimental group studying the B and T cell repertoire in autoimmunity. ? ? ?
