Human Pancreas Atlas Explorer
Interactive exploration of scRNAseq and spatial transcriptomics datasets
Study Overview
These interactive applications enable comprehensive exploration of single-cell RNA sequencing (scRNAseq) and spatial transcriptomics datasets from healthy donor pancreas and pancreatic ductal adenocarcinoma (PDAC) samples. Data are integrated from Elhossiny et al., 2025, Steele et al. (2020), Carpenter et al. (2023), and Carpenter et al. (2024) studies.
The applications are built using ShinyCell2 (Chen et al. 2025) with custom modifications. Seurat objects of the integrated datasets will be available upon publication.
If you use these applications in your research, please cite the original studies listed above.
Applications
Explore ~200,000 cells from healthy donor pancreas and PDAC samples with cell type annotation and gene expression visualization.
Explore spatial transcriptomics data of 14 different samples from healthy donor pancreas and PDAC with paired H&E histology images
Explore ~160,000 Visium spots across healthy donor and PDAC spatial samples with spatial domain annotation and gene expression visualization.
User Guide
For optimal viewing and interaction, we recommend using these applications on a desktop or laptop computer with a modern web browser.
Interface Overview
Each application features an interactive sidebar for customizing visualizations:
scRNAseq Atlas & Integrated Spatial Atlas Explorers
These applications provide comprehensive tools for integrated dataset exploration:
Interactive UMAP representation with zoom functionality to visualize cell types and gene expression patterns at multiple scales.
Simultaneous visualization of two genes or gene expression alongside cell type annotations for comparative analysis.
Violin plots and box plots showing gene expression across cell types and spatial domains, with built-in statistical testing capabilities.
Comparative visualization of expression patterns for multiple genes of interest.
Simultaneous visualization of two-gene co-expression patterns on UMAP representations.
Analysis of cell type proportions across different sample groups and conditions.
Spatial Samples Visualizer
This application enables detailed exploration of individual spatial transcriptomics samples with corresponding H&E Visualization, Full resolution images will be available upon publication.
Spatial mapping based on deconvolution results.
Visualization of defined tissue regions and microenvironments.
Methods
Data processing and analysis methods will be available upon publication. All analysis code and pipelines are publicly available in our GitHub repository.
Contact & Support
For questions, feedback, or collaboration inquiries, please contact:
Ahmed M. Elhossiny, PhD Study Author hossiny@umich.edu
Marina Pasca di Magliano, PhD Principal Investigator marinapa@umich.edu