DyMoTree decodes early cell state transitions and drivers from single-cell transcriptomes using a tree-structured neural network
Abstract
Inferring early cell fate from single-cell RNA-sequencing data is essential for identifying cellular origins and fate plasticity in development and disease. However, existing methods often fail to exploit tree-structured lineage trajectories, limiting the accuracy and interpretability of fate mapping. Here we present DyMoTree, a computational framework that models cell fate decisions as nonlinear mappings between progenitor and terminal cell states under explicit lineage constraints. By integrating lineage graphs with a tree-structured neural architecture, DyMoTree learns lineage-resolved cell-state transition maps from single-cell transcriptomes, enabling robust inference of early fate bias and identification of fate-specific progenitor substates and driver genes. Across simulations, lineage-tracing experiments, and in vivo systems, DyMoTree outperformed existing methods in resolving early fate biases. Applications to mouse embryogenesis, lung adenocarcinoma progression, and CAR-T immunotherapy revealed regulatory programs underlying developmental and disease-associated transitions. DyMoTree provides a general framework for modeling lineage-resolved cell-state dynamics underlying development and disease progression. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=156 SRC="FIGDIR/small/731114v2_ufig1.gif" ALT="Figure 1"> View larger version (49K): org.highwire.dtl.DTLVardef@11e715dorg.highwire.dtl.DTLVardef@1a4c695org.highwire.dtl.DTLVardef@e98805org.highwire.dtl.DTLVardef@1e14054_HPS_FORMAT_FIGEXP M_FIG C_FIG
Full-text reasoning
From the deep-tier full-text analysis of this preprint.
Reasoning review — 8 key claims, 1 well-supported, 7 with gaps
- supported DyMoTree models cell fate decisions as nonlinear mappings between progenitor and terminal cell states under explicit lineage constraints.
- partial DyMoTree enables robust inference of early fate bias, identification of fate-specific progenitor substates, and discovery of lineage-specific driver genes. gap: The claim of "robustness" for inference and identification is stated but not explicitly supported with evidence of resilience to noise or variability in the provided text.
- partial DyMoTree outperformed existing methods in resolving early fate biases across simulations, lineage-tracing experiments, and in vivo systems. gap: The specific methods compared, metrics of outperformance, and the extent of the improvement are not detailed in the excerpt.
- partial DyMoTree accurately recovered early fate biases in progenitor states across various experimental conditions. gap: No specific evidence or quantitative metrics are provided in the excerpt to demonstrate the claimed accuracy.
- partial DyMoTree successfully reconstructed dynamic expression trends of key regulators during the second cell-fate decision in mouse embryogenesis. gap: The claim is qualitative; no specific reconstructed trends or validation metrics are provided in the excerpt.
- partial DyMoTree inferred biologically interpretable transition patterns and recovered molecular mechanisms in LUAD and CAR-T immunotherapy datasets. gap: The claims are qualitative; no specific examples of interpretable patterns or recovered mechanisms are provided in the excerpt.
- partial DyMoTree is a general framework for modeling lineage-resolved cell-state dynamics underlying development and disease progression. gap: While diverse applications are listed, the claim of being a "general framework" might be an over-generalization given the limited number of specific applications detailed in the excerpt.
- partial DyMoTree identified a greater number of fate-biased HSPCs, particularly in monocyte/neutrophil branches, often with corresponding lineage barcodes. gap: The significance of identifying a "greater number" and the validation provided by "corresponding lineage barcodes" are not fully explained or quantified in the excerpt.
Claims and gaps are read from the full text by a language model, shown for transparency; they do not affect ranking or selection.
Lifecycle
Discussion
Moderated digest of third-party discussion on Bluesky — substantive endorsement and critique.
-
#SingleCell preprints @prepub-singlecell.bsky.social · 721 followers neutral
DyMoTree decodes early cell state transitions and drivers from single-cell transcriptomes using a tree-structured neural network #SingleCell 🧪🧬🖥️ https://www.biorxiv.org/content/10.64898/2026.06.09.731114v1
-
bioRxiv Bioinfo @biorxiv-bioinfo.bsky.social · 4920 followers neutral
DyMoTree decodes early cell state transitions and drivers from single-cell transcriptomes using a tree-structured neural network https://www.biorxiv.org/content/10.64898/2026.06.09.731114v1
-
AI x Bio Discovery @aixbiobot.bsky.social · 787 followers neutral
DyMoTree decodes early cell state transitions and drivers from single-cell transcriptomes using a tree-structured neural network [new]
-
bioRxivpreprint @biorxivpreprint.bsky.social · 8895 followers neutral
DyMoTree decodes early cell state transitions and drivers from single-cell transcriptomes using a tree-structured neural network https://www.biorxiv.org/content/10.64898/2026.06.09.731114v1