scphylo.datasets.colorectal1#

scphylo.datasets.colorectal1()[source]#

Human Colorectal Cancer (Patient 1).

This dataset was introduced in [Leung_2017] and was used in:

The size is n_cells × n_muts = 178 × 16

Returns

An anndata in which .X is the input noisy.

Return type

anndata.AnnData

Notes

This dataset includes single cells from two sites of the patient body; 133 single cells from colon as primary tumor site and 45 single cells from liver as the tumor metastatic site (178 in total). The number of mutations in this dataset is 16. One can remove the cells in this dataset that carry none of these 16 mutations before feeding it to our network. After removing cells with zero profile, the number of cells are 40 and 32 from primary and metastatic sites, respectively (72 in total).