Analyzing and Predicting Antibody Responses

The ability to predict and correspondingly manipulate an adaptive immune response would be highly valuable for biotechnology and medicine. To achieve this requires a greater molecular understanding of adaptive immunity. Recent advances in high-throughput immunoglobulin (antibody) repertoire sequencing (Ig-seq) are enabling highly quantitative analysis of adaptive immune responses. This increased immunological insight has been applied to fields as varied as lymphocyte development and differentiation, immunodiagnostics discovery, vaccine development, cancer immunotherapy, and monoclonal antibody discovery. Here I will present, the computational and experimental methods my group has developed in Ig-seq and their associated applications. For, example we have recently developed a comprehensive error and bias correction method that enables highly accurate Ig-seq. We are also applying machine-learning approaches that enable prediction of the immune status of individual antibody clones. Finally, in the context of adaptive immunity, we addressed the age-old question of the balance between nature, nurture and noise. It has remained unclear to what extent nature (genetic background) and nurture (antigen challenge) predetermine an antibody response and what fraction is private to an individual (noise). By employing ultra-deep Ig-seq (400 million reads) in mice, we discovered that that nature and nurture dictate up to 99% of the development of immunoglobulin repertoires. This is in contrast to current immunological dogma, which suggests that an antibody response is overwhelmingly unpredictable. The resolution of this question resolves a decisive blank area in the understanding and prediction of adaptive immunity.