Journal: bioRxiv
Article Title: 4D multimodal wound healing atlas reveals organ-level controls of repair phase transitions
doi: 10.64898/2026.01.15.699736
Figure Lengend Snippet: (A) Schematic overview of WoundScape, the organ-scale spatial transcriptomic wound atlas. High-resolution 2 μm Visium HD data were generated for UW, D7PW, D15PW, and D30PW skin, totaling 532,066 spatial barcoded spots. These data were merged with the OWHA omnibus, yielding a comprehensive tetra-modal spatially resolved database encompassing 725,590 total cells and spots organized across all anatomical compartments of the skin. Histological H&E staining was combined with WoundScape spatial profiling to precisely align Visium HD-identified neighborhoods within defined cutaneous wound anatomical regions. (B–E ) Spatial transcriptomic mapping of Banksy neighborhoods (bottom), and corresponding H&E sections (above), in 8 μm Visium HD sections from unwounded ( B ) (UW), ( C ) D4PW, ( D ) D7PW, and ( E ) D30PW. Middle insets show magnified views of local BANKSY neighborhoods clustering (clustering resolution = 0.5). Right insets depict corresponding RCTD-derived metacluster annotations for all discrete spatial domains within each section. Each section represents a technical replicate from the same biological specimen (UW = 113,587 spots; D4PW = 104,247; D7PW = 176,256; D30PW = 137,976). Red arrowheads indicate the initial wound edge in the suprabasal layer. White guidelines mark the initial subcutis wound boundaries. Scale bars represent 500 um or 1 mm as indicated. (F–I) Stacked bar plots showing the proportional metacluster composition within each BANKSY cluster for unwounded skin and wounded mouse skin (D4PW, D7PW, D30PW). Data represents two technical replicates derived from the same biological sample. Statistical similarity of metacluster compositions between replicates was assessed using a chi-square test with Monte Carlo permutation (100,000 simulations). BANKSY clusters that show significant concordance between technical replicates ( P < 0.05) are highlighted in red font. (J) Visium HD localization of Dominant Signalers and Central Orchestrators, including Basal IV, Papillary II, HF I, Myofibroblasts II, and Muscle Progenitor clusters, along BANKSY neighborhoods proximal to the wound bed at D7PW. Red outlines mark wound-associated regions of interest, while yellow guidelines and red arrowheads denote suprabasal and subcutis wound edge boundaries, respectively. Scale bars, 1 mm. (K) Anatomical distance quantifications of wound emergent BANKSY clusters along the anterior-posterior transverse plane at D7PW. The x-axis represents arbitrary spatial units (1 a.u. = 9 mm) corresponding to anterior-posterior distance across the entire tissue section. Vertical redline demarcates the wound center. Data represents two technical replicates from the same biological timepoints. A variance-based localization test was used, and multiple testing correction was applied to p-values using the Benjamini–Hochberg procedure. (p < 0.05 = *, p < 0.01 = **, p < .001 = ***). (L) Schematic illustration summarizing the spatial geometry of the wound edge (anterior and posterior margins) as visualized at D7PW. Red arrowheads indicate typical location of wound boundaries from the adjacent wound bed. (M-N) High-magnification 20x H&E of the D7PW wound edge regions of interest (ROI). Red arrowheads indicate wound boundaries. Scale bars, 500 μm. (O-T) BANKSY spatial clustering of respective Dominant Signalers populations within the posterior wound edge ROI: ( O ) BANKSY cluster positions, ( P ) merged overlay of selected Dominant Signaler populations: ( Q ) Spinous I, ( R ) Proliferative Endothelial Cells, ( S ) Pericyte I, and ( T ) HF I. Each overlay highlights discrete but spatially organized domains at the wound front where reparative signaling networks converge. Scale bars, 100 μm. (U) Stacked bar plots showing fine cell-type composition of BANKSY clusters localized at the D7PW wound front. Data represents two technical replicates from the same biological timepoints. (V) Numbering and classification legend of fine cell types corresponding to panel (U) .
Article Snippet: Spatial Transcriptomics data was generated using 10x Genomics Visium V2 CytAssist Spatial Gene Expression Mouse Transcriptome Assay (#1000445) for FFPE tissue as per user’s guide.
Techniques: Generated, Staining, Derivative Assay
