Project Abstract T cell recognition of peptide-Major Histocompatibility Complexes is a key determinant of response to infection, cancer, and autoimmunity. While there have been recent technological advances that have enabled better tracking and analysis of the T cell receptor repertoire, these approaches are extremely resource intensive and/or have key technical limitations. Here, we propose a novel method that combines emulsion-based partitioning and computational sequence deconvolution to enable large-scale determination of the naive T cell repertoire at modest cost (estimated to be ~$1/3,000 cells at current reagent and sequencing costs, as compared to the ~$1/cell for current techniques). We will first develop and validate the experimental modifications to establish this technique, including development of low-cost DNA barcoding beads, formation of cell/bead emulsions, and efficient conversion and capture of TCR transcripts. We will then extend our previous computational approach to deconvolute pools of TCRα/TCRβ sequences, and validate our method on T cells obtained from healthy donors. In addition, we will extend our methodology to include oligonucleotide tagged pMHC multimers, enabling repertoire-scale tracking of TCRα/TCRβ-pMHC pairings. Together, these technologies will enable efforts to track the T cell repertoire at extremely large scale, an advance necessary for both mechanistic immunology and computational prediction of antigen reactivity.