Autoimmune disease burdens an estimated 10% of the human population. Environmental factors have long been suspected to initiate, drive, or otherwise contribute to autoimmune disease development and progression. In spite of tremendous effort the identification of specific organisms involved has proven exceptionally difficult.
We have developed a separation process framework to identify those particular antibodies in human blood that associate with a specific disease, and to identify their preferred molecular specificity. In this way, we can map the antibody binding specificity to a small set of candidate protein antigens from the environment.
This process involves the identification of individual peptides present in a library of 10 billion random sequences that bind an antibody present in the diverse repertoire of an individual patient. The sequences of the binding peptides are then determined using massively parallel (or “next-generation”) DNA sequencing. Next, a novel computational algorithm was developed to identify antibody-binding patterns or motifs present in data sets containing tens of millions of unique sequences. Finally, individual motifs are experimentally validated and refined using molecular evolution to aid the identification of specific antigens from the environment that associate with disease.
This approach offers the prospect of enabling the identification of specific organisms and immune responses that directly cause or contribute to human autoimmune diseases.