1
Integrated multi-center atlas
Unified single-cell and spatial data from multiple international PDAC cohorts.
Pancreatic cancer systems atlas
PANCOSMOS
PANcreatic Cancer: an Organized Single-cell Map Of Spatial ecosystems
An integrated atlas connecting cellular states with spatial tissue ecosystems in pancreatic ductal adenocarcinoma.
>1 million single cells and nuclei
125 spatial sections
Search gene expression
Explore gene expression across cell types and spatial modules.
One atlas, two scales of tissue organization.
PANCOSMOS connects cellular identity with spatial tissue architecture across pancreatic ductal adenocarcinoma. Explore conserved cell states, compare multicenter cohorts, map spatial neighborhoods, and relate tumor ecosystems to clinical outcome from one coordinated resource.
>1M
Integrated single cells and nuclei
275
Tumor and adjacent-normal specimens
110
Spatial transcriptomic sections
14
Spatial modules defining tissue structure
01
Single-cell landscape
An integrated view of malignant, immune, stromal, vascular, neural, and epithelial cell states.
02
Spatial tissue architecture
Cellular neighborhoods and spatial modules in tissue context.
02What the atlas lets you do
2
Million-cell landscape
Over one million single cells defining PDAC tumor microenvironment heterogeneity.
3
AI-ready resource
High-quality dataset optimized for virtual cell modeling and AI training.
4
Transferable cell annotation
CellTypist-based transfer learning for consistent, hierarchical labels across datasets.
5
Gene expression explorer
Query and visualize gene expression across single-cell and spatial datasets.
6
Spatial colocalization network
Map spatial organization and cell–cell proximity interactions within tissues.
7
Survival analysis module
Gene- and gene-set–level survival modeling across 11 bulk cohorts.
8
FAIR data access
Standardized, accessible, and citable data following FAIR principles.
Collection index
EXPLORE THE ATLASRESEARCH LENSES / 03
Move from a gene to a tissue ecosystem.
Start with a gene, cell population, or spatial module. PANCOSMOS provides linked views for expression, cellular composition, spatial proximity, and survival associations so that observations can be followed across biological scales.
A / CELL STATES
Resolve malignant and microenvironmental programs
Compare basal-like, classical-like, exocrine-like, and NRP-like tumor programs alongside immune and stromal populations.
B / SPATIAL ECOLOGY
Trace how cell states assemble in tissue
Examine spatial modules, colocalization patterns, immune niches, and ecotypes across Visium and Stereo-seq sections.
Single-cell gene query
Enter one or more genes. Separate multiple genes with commas, semicolons, spaces, or line breaks.
Spatial-module gene query
FC across spatial modules; bubble size represents the observed proportion.
Survival Analysis
Download Center
Single-Cell
Integrated level-4 single-cell RDS objects (lineage-specific). Click to download:
- PANCOSMOS_Lymphoid_B_IntegratedscRNA.rds (291 MB)
- PANCOSMOS_Lymphoid_NK_IntegratedscRNA.rds (107 MB)
- PANCOSMOS_Lymphoid_T_IntegratedscRNA.rds (1.02 GB)
- PANCOSMOS_Myeloid_DC_IntegratedscRNA.rds (95.8 MB)
- PANCOSMOS_Myeloid_Macrophage_IntegratedscRNA.rds (487 MB)
- PANCOSMOS_Myeloid_Mast_IntegratedscRNA.rds (87 MB)
- PANCOSMOS_Myeloid_Neutrophil_IntegratedscRNA.rds (55 MB)
- PANCOSMOS_Parenchymal_Endocrine_IntegratedscRNA.rds (180 MB)
- PANCOSMOS_Parenchymal_Exocrine_IntegratedscRNA.rds (3.54 GB)
- PANCOSMOS_TumorCell_IntegratedscRNA.rds (2.06 GB)
- PANCOSMOS_Stroma_Endothelial_IntegratedscRNA.rds (297 MB)
- PANCOSMOS_Stroma_Fibroblast_IntegratedscRNA.rds (1.05 GB)
- PANCOSMOS_Stroma_Schwann_IntegratedscRNA.rds (26.6 MB)
Stereo-seq level3 bin100 Seurat objects
Processed Seurat objects (bin100). Click to download:
- PUMCH_1_bin100_seuobj.rds (135 MB)
- PUMCH_2_bin100_seuobj.rds (60.2 MB)
- PUMCH_3_bin100_seuobj.rds (81.3 MB)
- PUMCH_4_bin100_seuobj.rds (50.7 MB)
- PUMCH_5_bin100_seuobj.rds (44.2 MB)
- PUMCH_6_bin100_seuobj.rds (107 MB)
- PUMCH_7_bin100_seuobj.rds (92.5 MB)
- PUMCH_8_bin100_seuobj.rds (107 MB)
- PUMCH_9_bin100_seuobj.rds (251 MB)
- PUMCH_10_bin100_seuobj.rds (229 MB)
- PUMCH_11_bin100_seuobj.rds (130 MB)
- PUMCH_12_bin100_seuobj.rds (81.7 MB)
- PUMCH_13_bin100_seuobj.rds (39.6 MB)
- PUMCH_14_1_bin100_seuobj.rds (36.7 MB)
- PUMCH_14_2_bin100_seuobj.rds (38.9 MB)
- PUMCH_15_bin100_seuobj.rds (59.5 MB)
Stereo-seq (bin1)
Raw GEM files for PUMCH (bin = 1). Click to download:
- PUMCH_1.gem (948 MB)
- PUMCH_2.gem (424 MB)
- PUMCH_3.gem (614 MB)
- PUMCH_4.gem (257 MB)
- PUMCH_5.gem (216 MB)
- PUMCH_6.gem (577 MB)
- PUMCH_7.gem (651 MB)
- PUMCH_8.gem (670 MB)
- PUMCH_9.gem (2.13 GB)
- PUMCH_10.gem (1.91 GB)
- PUMCH_11.gem (999 MB)
- PUMCH_12.gem (521 MB)
- PUMCH_13.gem (270 MB)
- PUMCH_14_1.gem (239 MB)
- PUMCH_14_2.gem (245 MB)
- PUMCH_15.gem (427 MB)
CellTypist cell type prediction models
Pre-trained models for lineage-level prediction. Click to download:
- PANCOSMOS_PDAC_TME_celltype_Level1.pkl (822 KB)
- PANCOSMOS_PDAC_B_Lineage.pkl (150 KB)
- PANCOSMOS_PDAC_DendriticCell.pkl (113 KB)
- PANCOSMOS_PDAC_EndocrineCell.pkl (86.8 KB)
- PANCOSMOS_PDAC_EndothelialCell.pkl (97.2 KB)
- PANCOSMOS_PDAC_ExocrineCell.pkl (107 KB)
- PANCOSMOS_PDAC_FibroblastPSC.pkl (165 KB)
- PANCOSMOS_PDAC_Innatelymphoid.pkl (74.3 KB)
- PANCOSMOS_PDAC_Macrophage.pkl (161 KB)
- PANCOSMOS_PDAC_MastCell.pkl (13.1 KB)
- PANCOSMOS_PDAC_Neutrophil.pkl (13 KB)
- PANCOSMOS_PDAC_SchwannCell.pkl (45 KB)
- PANCOSMOS_PDAC_T_Lineage.pkl (162 KB)
- PANCOSMOS_PDAC_TumorCell.pkl (166 KB)
python
import celltypist
pred = celltypist.annotate(adata, model='PDAC_TME_celltype_Level1.pkl')
pred.predicted_labels