patpy.datasets.ticatlas

Contents

patpy.datasets.ticatlas#

patpy.datasets.ticatlas(kind='processed', overwrite=False, return_dataset_info=False)#

Tumor Immune Cell Atlas (TICAtlas) dataset.

The processed version was prepared with the standard scanpy pipeline; cells annotated as “nan” were removed; PCA, scVI, scANVI, and scPoli dimensionality reduction were applied. The dataset contains 267,547 cells and 3,000 features. The processed download is approximately 0.5 GB compressed and ~1.8 GB unzipped.

Parameters:

  • kind (Literal['raw', 'processed'] (default: 'processed')) – Either "processed" (default) or "raw". Currently only "processed" is available; "raw" raises NotImplementedError.

  • overwrite (bool (default: False)) – If True, re-download the dataset even when a cached copy exists.

  • return_dataset_info (bool (default: False)) – If True, return a tuple (adata, DatasetInfo) instead of just adata.

References

Nieto, P., Elosua-Bayes, M., Trincado, J. L., Marchese, D., Massoni-Badosa, R., Salvany, M., … & Heyn, H. (2021). A single-cell tumor immune atlas for precision oncology. Genome research, 31(10), 1913-1926. https://doi.org/10.1101/gr.273300.120

Return type:

AnnData | tuple[AnnData, DatasetInfo]

Returns:

AnnData object of scRNA-seq profiles, optionally paired with a DatasetInfo describing the dataset’s standard schema.

Examples

>>> import patpy
>>> adata = patpy.datasets.ticatlas()
>>> adata, info = patpy.datasets.ticatlas(return_dataset_info=True)
Contents