Uni-TINT: Unifying T and Innate Lymphoid Cell Taxonomy with a high resolution, pan-disease and pan-tissue single-cell transcriptomic atlas

Abstract

T cells and innate lymphoid cells (ILCs) exhibit extensive phenotypic diversity across tissues and diseases, yet inconsistent annotation limits cross-study comparisons and biological interpretation. We present Uni-TINT, an integrated pan-disease and pan-tissue atlas comprising 3.46 million cells from 1,869 samples spanning 194 studies, 166 disease subtypes, and 66 tissue types. Through systematic, hierarchical manual annotation, Uni-TINT establishes a unified and context-aware taxonomy of T and ILC populations, resolving 207 cell subtypes and states across conventional and unconventional T cells, natural killer (NK) cells, helper ILCs, thymocytes, and hematopoietic progenitors. We identified an immunosuppressive, tumour-associated CD4 T regulatory population which we validated with spatial transcriptomics. Other rare and unconventional populations characterised included CD8 regulatory T cells, invariant NKT cells and memory-like NK cells. Integration of T cell receptor sequencing suggested functional associations between {gamma}{delta} T cell co-receptor expression and TRDV gene usage. Finally, a comparative analysis of healthy and diseased immature cells identified a small population of malignant hematopoietic stem cells carrying chromosomal aberrations and enriched in acute leukaemia of mixed phenotype. Together, Uni-TINT provides a unified reference framework for immune annotation and discovery across health and disease.