Human pancreas

PDAC-SMI and co-culture dataset (T4-a)

SpatialData object
├── Shapes
│     ├── 'cell_circles': GeoDataFrame shape: (56312, 2) (2D shapes)
│     └── 'fov_boxes': GeoDataFrame shape: (21, 2) (2D shapes)
└── Tables
      └── 'table': AnnData (56312, 1009)
with coordinate systems:
    ▸ 'global', with elements:
        cell_circles (Shapes), fov_boxes (Shapes)
    ▸ 'microns', with elements:
        fov_boxes (Shapes)

from sddb import Catalog

ds = Catalog().dataset("rxu3X0FD6GTLBiDm0000")
ds.elements()          # inspect without downloading
sdata = ds.to_sdata()  # cached locally, reused
print(sdata)

import lamindb as ln

db = ln.DB("scverse/spatialdata-db")
artifact = db.Artifact.get("rxu3X0FD6GTLBiDm")   # stem uid: Lamin resolves it to the latest version

sdata = artifact.load()   # cache, then load into memory
group = artifact.open()   # stream without downloading
path  = artifact.cache()  # sync into the local cache, returns the path

# sddb client: download the .zarr into data/
sddb-cli download rxu3X0FD6GTLBiDm0000 --dest data/

# or lamin: cache the dataset and print its local path
lamin load https://lamin.ai/scverse/spatialdata-db/artifact/rxu3X0FD6GTLBiDm

# or connect once, then load by uid
lamin connect scverse/spatialdata-db
lamin load artifact --uid rxu3X0FD6GTLBiDm

Metadata

Technology

CosMx SMI

Vendor

Bruker

Organism

human

Tissue

pancreas

Condition

pancreatic ductal adenocarcinoma

Modality

RNA

Observations (cells)

56,312

Target genes

990

Tier

bronze

Size

41.6 MB

Sample ID

kx6b69n3cb:T4-a

License

CC BY 4.0

Published

Jul 11, 2024

Review

raw_meta_data_final.h5ad header (5 MB read 2026-11-05): X csr counts, obs sample_id, treatment_status, donor, fov, CenterX/Y_global_px; treatment_status values not decoded (legacy categories); tumour group from the folder prefix: T = neoadjuvant-treated, U = treatment-naive is inferred (description says 13 tumours, treated or naive; 6 T + 7 U ids match 13)

@article{mendeley:kx6b69n3cb,
  doi = {10.17632/kx6b69n3cb.1},
  title = {PDAC-SMI and co-culture dataset (T4-a)},
  year = {2024}
}