High-throughput, quantitative approach to peptide epitope discovery

BACKGROUND AND OBJECTIVE: PCR-based, immune repertoire data is commonly used to assess disease features, but such data has not yet been used to discover epitopes. METHODS: This report represents the development of an algorithm that utilizes IGH CDR3s, along with adaptive immune receptor, antigen chemical complementarity algorithms, to identify candidate epitopes within known antigens. A ratio that accounts for the number of times each IGH CDR3 within an immune repertoire sample complements a particular amino acid (AA) residue and the number of unique individual IGH CDR3s that complement that same residue was obtained to initiate this IGH CDR3-epitope matching algorithm. Then, these ratios, representing each antigen AA, were weighted by the size of the immune repertoire samples. This process allowed a comparison to a collection of control immune repertoire samples, whereby IGH CDR3s representing high diversity, chemical complementary, and frequency effectively identified epitope candidates. RESULTS: The indicated algorithm was successful in the de novo identification of several established epitopes for multiple sclerosis and celiac disease, respectively, and in the de novo identification of other candidate epitopes. Also, the above algorithm identified similar patterns of IGH CDR3 diversity and chemical complementarity, but not similar IGH CDR3 frequencies, to known disease epitopes among healthy controls, possibly indicating a basis for a human predisposition to autoimmunity. CONCLUSION: This report indicates the opportunity for computational epitope discovery; and for the computational monitoring of distinct, patient adaptive immune receptor-antigen reactivities.