Transcriptional landscape of direct reprogramming toward hematopoietic stem cells

Direct reprogramming of human fibroblasts into hematopoietic stem cells (HSCs) offers a promising strategy for generating autologous cells to treat blood and immune disorders. How-ever, low conversion efficiency and incomplete characterization of reprogrammed states limit current protocols. We therefore assembled a single-cell transcriptomic reference atlas spanning hematopoietic, stromal, and engineered cell types for evaluating reprogramming outcomes, and tested an algorithmically-predicted transcription factor recipe—GATA2, GFI1B, FOS, REL, and STAT5A—for HSC induction. Long-read single-cell RNA sequencing of CD34 + cells re-vealed heterogeneous reprogramming intermediates expressing fibroblast-, endothelial-, and hematopoietic-associated genes, consistent with partial reprogramming. We developed a quan-titative benchmarking approach by mapping reprogrammed cells’ transcriptomic positions to reference atlas cell types, revealing an endothelial-to-hematopoietic transition-like trajectory. Finally, we cataloged isoform diversity in reprogrammed cells, initial fibroblasts, and target HSCs, demonstrating transcriptional remodeling that is not apparent from gene-level analy-ses alone. This experimental-computational framework provides a generalizable strategy for characterizing and optimizing reprogramming protocols.