LOTUS: lineage-aware optimal transport reconstructs continuous clonal fate dynamics

LOTUS reconstructs continuous clonal fate dynamics from time-resolved single-cell snapshots by enforcing lineage-consistent optimal transport, reducing off-target routing from 4.94–14.84% to 1.10% in branching simulations.

• Why it matters: Destructive single-cell sampling breaks ancestor–descendant links, making it hard to recover continuous cell-fate trajectories and clonal propagation from discrete snapshots.
• What they did: LOTUS integrates time-resolved transcriptomic snapshots with clonal lineage from engineered barcodes or endogenous T cell receptor clonotypes, coupling clone-level optimal transport with neural unbalanced transport under lineage-consistent constraints.
• The result: In LARRY hematopoiesis, it reconstructs branching trajectories and improves early neutrophil–monocyte fate-boundary prediction; in CellTagging fibroblast reprogramming, it separates reprogrammed from failed trajectories; in malaria-responsive memory CD45RA− CD4+ T cells, it preserves clonotype structure and reconstructs TR1 dynamics and regulatory networks.

AI summary of the abstract · may contain errors · not medical advice · read the paper.