Mapping the Landscape of CD8+ T Cells in Inflammation and Cancer

Researchers have unveiled an extensive atlas of human CD8+ T cells, showcasing their diversity in inflammation and cancer using the novel deep-learning model, scAtlasVAE. This tool integrates large-scale single-cell RNA sequencing data, encompassing over 1.1 million cells from 961 samples across 42 disease conditions. By incorporating T cell receptor clonal information, the study connects cell subtypes and highlights their phenotypic and functional transitions. Notably, it identifies three distinct exhausted T cell subtypes and reveals their unique transcriptomic and clonal sharing patterns in autoimmune conditions and immune-related adverse events. The scAtlasVAE framework also enables automated and scalable annotation of CD8+ T cell subtypes, providing a powerful resource for unbiased single-cell analyses and advancing our understanding of T cell dynamics in health and disease.