PROJECT SUMMARY/ABSTRACT Type 1 diabetes (T1D) is a serious multi-factorial chronic autoimmune disease with an annual 3% increase in the incidence rate that constitutes a major public health challenge and financial burden. T1D involves genetic predisposition, immune system response, and environmental factors that lead to disease initiation and progression. Due to the lack of curative therapies for T1D, the most promising option to date remains early intervention with the goal of slowing or preventing progression to T1D in predisposed individuals. Current diagnosis of the pre-clinical T1D stage is based on the detection of islet autoimmunity (IA) against two or more specific antigens, i.e. when the destruction of β-cells has already started and is difficult to reverse. A diagnostic tool predicting the development of islet autoantibodies early in the progression has the potential to avoid the destruction of β-cells altogether by using primary prevention strategies. Here, it is hypothesized that prior to the development of IA there is a distinct humoral immune response against immunogenic pathogen-specific and/or associated non-islet autoimmune targets that can be utilized as early risk stratification for progression to IA. The proposed approach relies on representing an entire binding space of a donor’s circulating antibody repertoire using machine learning models based on the antibody binding profile to a diverse, random library of 126,050 peptides with an average length of 9 amino acids, which is a sparse representation of all possible amino acid combinations. Resulting models are then used to identify pathogen epitopes with high predictive power that are combined into a panel with diagnostic efficacy. The overarching goal of this study is to develop a panel of biomarkers, consisting of potential viral antigens and autoimmune targets for early prediction of islet autoimmunity in genetically susceptible individuals. A broad profiling of the circulating antibody repertoire in patient’s serum combined with machine learning models over time will be used to discover immunogenic targets in both pathogen and human proteomes that can be used as predictors of progression to IA and T1D. The serologic (autoantibody detection), genetic (HLA genotype, point mutations) and clinical data will be used in combination with the immune response profiling data to investigate temporal alterations in humoral immune response at different timepoints of progression to IA. This work is expected to yield data demonstrating the feasibility of a novel immunoassay for early risk stratification of islet autoimmunity development in genetically predisposed T1D individuals. Additionally, it will serve as a demonstration of the antigen discovery approach as a means to identify diagnostic antigens for difficult pathogens.