this repository is archived as this functionality is now in cellxgene
CLI for preprocessing data for use with cellxgene
Prepare a dataset example.h5ad
by calling
cellxgene-prepare example.h5ad --engine=scanpy --output=example_processed.h5ad
This will save a new file example_processed.h5ad
with a few new fields prepared designed for use with the specified engine.
Several options are available for specifying preprocessing, generating and saving plots, ensuring data sparsity, and others, to see all options call
cellxgene --help
Options:
--engine [scanpy] computational engine [default: scanpy]
--format [h5ad|10x_mtx|loom] data format [default: h5ad]
--layout [umap|tsne|umap+tsne] layout algorithm [default: umap]
--recipe [none|seurat|zheng17] preprocessing receipe to run [default:
none]
--output TEXT whether or not to save a new file
--sparse / --no-sparse whether to require sparsity
--plotting / --no-plotting whether or not to generate plots
--help Show this message and exit.
For development and testing purposes, just install the requirements into your Python environment, and call
python prepare/cli.py example.h5ad
to simulate launching the CLI