Back

Trimodal Single-Cell Gene Regulatory Networks Reveal Principles of Stemness Loss and Cell Fate Acquisition in Human Hematopoiesis

Palii, C. G.; Tur, S.; Yan, S.; Longabaugh, W. J. R.; Solano, R.; Dilworth, F. J.; Ranish, J. A.; Brand, M.

2025-09-16 genomics
10.1101/2025.09.11.675740 bioRxiv
Show abstract

Hematopoiesis requires the coordinated loss of stemness and acquisition of lineage identity, yet the regulatory logic and principles underlying these transitions has remained elusive. Single-cell studies suggest that hematopoietic stem and progenitor cells progress along continuous trajectories, but this view conflicts with the existence of discrete, functionally validated populations. Here, we establish the first dynamic, enhancer-based gene regulatory networks (eGRNs) that resolve the molecular programs underlying early human hematopoietic fate decisions. Built on a high-resolution trimodal framework generated with TEA-seq, these networks integrate simultaneously measured transcription factor abundance, enhancer and promoter accessibility, and gene expression within single-cell trajectories anchored to immunophenotypically defined populations. Our framework reveals that stemness loss and lineage acquisition are temporally and mechanistically uncoupled: stemness programs decline gradually through reduced TF abundance long before chromatin closure, whereas lineage identity emerges stepwise through enhancer reconfiguration and activation of lineage-defining eGRNs. This process generates discrete regulatory states that align with immunophenotypically defined populations. Together, these findings reconcile continuous and discrete models of hematopoiesis and establish eGRNs as a powerful framework for defining cell types by their regulatory logic. In addition, we provide an interactive web-based resource to facilitate further investigation of eGRNs and trajectories during early human hematopoiesis.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.