Figure 1. Transcriptomic map of DBA⁺pancreatic cells.

(A) Schematic of experiment workflow. (B) The Uniform Manifold Approximation and Projection depicts murine pancreatic DBA⁺ cells obtained using the protocol. (C) A matrix plot shows average expression of ductal cell markers in all clusters, identifying clusters 0 and 8 as ductal cells. (D) Feature plots illustrate markers of various cell types including epithelial (Epcam), ductal (Krt19 and Sox9), CD45⁺ immune cells (Ptprc), endothelial cells (Pecam1), fibroblasts (Col1a1), endocrine cells (Chga), and acinar cells (Pnliprp1). We observed low-level expression of acinar cell markers uniformly across all clusters that is likely contaminating acinar cell mRNA.

Figure 1—figure supplement 1. Features of DBA⁺ (clusters 0–15) and ductal (clusters 0–5) cells.

(A) The sorting strategy for live, DBA⁺ pancreatic cells is displayed. The graph on the left shows the cells plotted by forward and side scatter. In this graph, red cells are PI⁺, purple cells are DBA⁺PI^-, and blue and green cells are doublets. The graph on the right shows the gating strategy used to sort DBA⁺PI^- pancreatic cells. (B) The number of genes and transcripts for each cell in clusters 0–15 is shown. (C) Co-immunofluorescence labeling identifies CK19⁺DBA⁺ and Collagen I⁺DBA⁺ copositive pancreatic cells and CD45⁺DBA⁺ copositive pancreatic lymph node cells. Pancreatic CD31⁺ cells are not DBA⁺. A yellow arrow points to a copositive cell. Scale bars are 50 µm. (D) The plot shows whether a ductal cell (from the 0–15 cluster dataset) is in cluster 0 or 8 in the 0–5 cluster Uniform Manifold Approximation and Projection (UMAP). (E) This plot depicts the location of the ductal clusters 0–5 in the 0–15 cluster UMAP. (F) Violin plots show the number of genes and transcripts in each cell for ductal clusters 0–5. (G) Feature plots depict expression of genes normally enriched in pancreatic duct cells.