Comprehensive multi-omics single-cell data integration reveals greater heterogeneity in the human immune system

Single-cell transcriptomics enables the definition of diverse human immune cell types across multiple tissue and disease contexts. Still, deeper biological understanding requires comprehensive integration of multiple single-cell -omics (transcriptomic, proteomic, and cell receptor repertoire). To improve the identification of diverse cell types and the accuracy of cell-type classification in our datasets, we developed SuPERR-seq, a novel analysis workflow to increase the resolution and accuracy of clustering and allow for the discovery and characterization of previously hidden cell subsets. We show that by incorporating information from cell-surface proteins and immunoglobulin transcript counts, we accurately remove cell doublets, and prevent widespread cell-type misclassification. This approach uniquely improves identification of heterogenous cell types in the human immune system, including a novel subset of antibody-secreting cells in the bone marrow. Overall design: scRNA-seq of the gene expression profile, surface-protein expression profile and the V(D)J profile of 3 Peripheral blood mononuclear cell (PBMC) and 2 Bone marrow (BM) from healthy donors. Please note that processed data was generated from all GEX, ADT and V(D)J raw data and is linked to the corresponding *[GEX] sample records.