scNT-seq human hematopoiesis

In our dynamo paper, we firstly highlight its power to overcome fundamental limitations of conventional splicing-based RNA velocity analyses to enable accurate velocity estimations on a metabolically labeled human hematopoiesis scRNA-seq dataset, which will be introduced first in this labeling scRNA-seq section. Under the differential geometry section, we will show how dynamo’s differential geometry analyses reveal mechanisms driving early megakaryocyte appearance, dual origin of basophil lineage and elucidate asymmetrical regulation within the PU.1-GATA1 circuit. Lastly, under the vector field predictions section, we will show we can leverage the least-action-path method to accurately predicts drivers of numerous hematopoietic transitions. Finally, in this section, we will also show how in silico perturbations predicts cell-fate diversions induced by gene perturbations. As you can, dynamo represents an important step in advancing quantitative and predictive modeling of cell-state transitions. What is also worth to mention is that although the demonstration here focuses on our labeling data, dynamo can be equally used to analyze conventional scRNA-seq (the splicing data) and even used in conjunctions with other velocity toolkits. It is just that some splicing data has poor results because splicing data has biased capture of introns and lack the real time scale because of unknown intrinsic splicing rate. Such limitations come from the data itself cannot in general be solved with computational models.

• Molecular mechanism of megakaryocytes
  • How to use dynamo to gain mechanistic insights with its unique differential geometry analyses
  • Map the topography of human hematopoiesis and associative fixed points
    • Associate fixed points to cell types
      • Lineage tree of hematopoiesis
  • Vector field pseudotime
  • Molecular mechanisms underlying the early appearance of the Meg lineage
    • High expression of FLI1, instead of KLF1, beginning at the progenitor state
    • High speed, acceleration and low divergence and curvature of the Meg lineage
    • A schematic diagram summarizing the interactions involving FLI1 and KLF1
• Minimal network for basophil lineage commitment
  • Regulatory network governing the Bas lineage’s dual origins
    • A minimal network governing GMP vs. Bas origin of Bas lineage
• Cell-wise analyses: indepth analyses of Pu.1 and GATA1 interactions
  • Three approaches for in-depth network motif characterizations
  • Cell-wise analyses of the PU.1/SPI1–GATA1 network motif across all cells
    • Response heatmap
  • Conclusion