Fig. 6: ClusterMap enables 3D in situ transcriptomics analysis.
From: ClusterMap for multi-scale clustering analysis of spatial gene expression
a Raw fluorescent signals of 3D STARmap cardiac organoid 8-gene dataset. Width: 465 µm, height: 465 µm, depth: 97 µm. b, c ClusterMap generates 3D cell segmentation map (b) and cell-type map (c) of (a), which includes 1519 cells. Insets in (a–c) show zoomed-in views of the highlighted regions. d ClusterMap generates a volumetric cell segmentation map of 3D STARmap mouse V1 28-gene dataset6, showing 24,590 cells. Width: 1545 µm, height: 1545 µm, depth: 100 µm. e The 3D cell type maps of d show the spatial cell type distribution. f The 3D tissue region map of (e). SC, subcortical. g Bar plots of composition of 11 cell types across 7 tissue regions (layers). h–j Example of cellular communication at a Pv, Sst, or Vip neuron, respectively. Left: schematics of 3D Delaunay triangulation of the Pv, Sst, or Vip neuron (highlighted in a white circle) and its first tier of neighboring cells. Middle: 3D spatial cell distribution of the first panel with the first tier of neighboring cells colored in white. Right: 3D spatial cell distribution of the first panel. Width 184 µm, height 194 µm, depth 100 µm. k, Bar plots of average composition of cell types around each cell type. Patterns of self-association in the minority inhibitory neurons are highlighted in the bounding box. Cell types in g–k are color-coded as in e. Data are presented as mean values ± SEM.
