Characterization of human leucocyte cell identity based on the expression of L1 and HERV elements 3719

Abstract Description Transposable Elements constitute up to 45% of the human genome and their expression has diverse roles in human biology, but current classifications of hematological cell identity overlook their contribution. In this study, we applied a computational method that we developed to quantify TE expression specifically in single-cell RNA-seq data. We profiled L1 and HERV transcription at single-locus resolution in PBMCs. We found cell-type-specific TE expression in PBMCs, and 28 TE transcripts -from families such as HARLEQUIN, MER4B, ERV316A3- and as markers for distinct cell types. Dendritic cells (DCs), despite being a minor PBMC population, expressed the most TEs, with an average of 19 HERV and 102 L1 features per cell. Notably, plasmacytoid dendritic cells (pDCs), known for their antiviral role, were the primary TE-expressing subset. We selected gene and TE transcripts with high biological cell-to-cell variation (i.e. highly variable features or HVFs) and performed clustering and dimensionality reduction. Using all HVFs (genes, HERVs, and L1s) distinguished major PBMC lineages and cell types. Unsupervised clustering solely based on HERV HVFs revealed subclusters within established cell types. These findings demonstrate the potential of TE expression profiles to uncover novel aspects of cell identity. Integrating TE expression into cell classification techniques enhances our understanding of immune cell heterogeneity and provides new insights into leukocyte biology. Funding Sources This study was funded in part by NIH Grants UM1AI164559 and R01CA260691 Topic Categories Computational and Systems Immunology (COMP)