Datasets

Datasets#

Convenience small datasets

spatialdata.datasets.blobs(length=512, n_points=200, n_shapes=5, extra_coord_system=None, n_channels=3, c_coords=None)[source]#

Blobs dataset.

Parameters:

  • length (int (default: 512)) – Length of the image/labels.

  • n_points (int (default: 200)) – Number of points to generate.

  • n_shapes (int (default: 5)) – Number of max shapes to generate. At most, as if overlapping they will be discarded

  • extra_coord_system (str | None (default: None)) – Extra coordinate space on top of the standard global coordinate space. Will have only identity transform.

  • n_channels (int (default: 3)) – Number of channels of the image

Return type:

SpatialData

Returns:

: SpatialData object with blobs dataset.

spatialdata.datasets.blobs_annotating_element(name)[source]#

Return the blobs dataset with the desired element annotated by the table.

Parameters:

name (Literal['blobs_labels', 'blobs_multiscale_labels', 'blobs_circles', 'blobs_polygons', 'blobs_multipolygons']) – Name of the element to annotate. One of “blobs_labels”, “blobs_multiscale_labels”, “blobs_circles”, “blobs_polygons”, “blobs_multipolygons”.

Return type:

SpatialData

Returns:

: SpatialData object with the desired element annotated by the table.

spatialdata.datasets.cells(path=None)[source]#

Cells dataset.

Download the cells example dataset and load it as a SpatialData object. The download is hash-verified and cached, so repeated calls reuse the local copy instead of downloading again.

The dataset is a small region of a Xenium Prime Cervical Cancer sample and contains three multiscale images (he_aligned, he_image, morphology_focus), three multiscale label layers (cell_labels, nucleus_labels, tissue_labels), the transcripts points, the cell_boundaries and nucleus_boundaries shapes, and a cell-by-gene table annotating the 94 cells.

Notes

Derived from the 10x Genomics Xenium Prime Cervical Cancer FFPE dataset (https://www.10xgenomics.com/datasets/xenium-prime-ffpe-human-cervical-cancer), subset to a small tissue region. Licensed under CC BY 4.0; see datasets.yaml for the attribution string shipped alongside the data.

Parameters:

path (str | None (default: None)) – Directory in which to cache the downloaded data. If None, the default OS cache location is used (pooch.os_cache() for "spatialdata").

Return type:

SpatialData

Returns:

: SpatialData object with the cells dataset.

spatialdata.datasets.raccoon()[source]#

Raccoon dataset.

Return type:

SpatialData