datasets.pbmc3k_processed().raw.to_adata() sc.pl.umap( pbmc, color="LGALS2", cmap=sns.cubehelix_palette(dark=0, light=.9, as_cmap=True) ) Or you could generate your own colormap where you can choose the color for a low value: sc.pl.umap( pbmc, color="LGALS2", # Felt ...
datasets.pbmc3k() sc.pp.filter_cells(example, min_genes=min_genes) sc.pp.filter_genes(example, min_cells=min_cells) example.X = example.X.toarray() example.raw = example sc.pp.normalize_total(example) example.layers['normalized'] = example.X.copy() sc.pp.log1p(example) example....
For the PBMC 10X data set (Supplementary Table S2), the standard normalization was taken, for the Drop-seq lung and the Smart-seq2 pancreas datasets the poscount normalization, which is more suitable for sparse data. Only cell types with at least 50 cells were analyzed to get a robust ...
Hi anndata team, I am trying to work with both raw count matrix and normed data. So I hope to do something like anndata.X = anndata.obsm['raw'].copy(). However, this process is super slow. I tried anndata.X.data = anndata.obsm['raw'].cop...
( settings.datasetdir / 'pbmc3k_processed.h5ad', backup_url='https://raw.githubusercontent.com/chanzuckerberg/cellxgene/main/example-dataset/pbmc3k.h5ad', ) def _download_visium_dataset( sample_id: str, spaceranger_version: str, base_dir: Optional[Path] = None, download_image: bool = ...