scanpy-cli

A command-line interface for Scanpy, a Python library for analyzing single-cell gene expression data.

Installation

pip install scanpy-cli

Usage

The scanpy-cli tool provides three main command groups for single-cell data analysis:

Preprocessing (pp)

Commands for preprocessing single-cell data:

• filter-cells: Filter cells based on counts or genes expressed
• filter-genes: Filter genes based on counts or cells expressing them
• normalize-total: Normalize counts per cell to a target sum (library-size normalization)
• log1p: Logarithmize the data matrix (log(X + 1))
• scale: Scale data to unit variance and zero mean
• calculate-qc-metrics: Compute per-cell and per-gene QC metrics
• downsample-counts: Downsample counts to equalize sequencing depth
• highly-variable-genes: Identify highly variable genes
• pca: Run principal component analysis
• neighbors: Compute neighborhood graph
• regress-out: Regress out unwanted sources of variation
• combat: Batch effect correction using ComBat
• harmony: Batch effect correction using Harmony
• bbknn: Batch-balanced k-nearest neighbor graph construction
• scanorama: Batch effect correction using Scanorama
• scrublet: Detect doublets in single-cell RNA-seq data

Tools (tl)

Commands for analysis tools:

• tsne: Run t-SNE dimensionality reduction
• umap: Run UMAP dimensionality reduction
• leiden: Run Leiden clustering
• paga: Run PAGA for trajectory inference
• rank-genes-groups: Find marker genes for clusters
• score-genes: Score a set of genes per cell

Plotting (pl)

Commands for visualization:

• umap: Plot UMAP embeddings

Input/Output (io)

Commands for reading and writing data:

• read-10x-h5: Read a 10x Genomics HDF5 file and save as .h5ad
• view: Display the structure of an .h5ad file (backed/read-only mode)

Development

Running Tests

To run the tests, install the package in development mode with test dependencies:

# Install in development mode with test dependencies
pip install -e ".[testing]"

# Or using uv
uv sync

# Run the tests with pytest
pytest

Code Quality

The project uses prek for code quality checks:

uv run prek run --all-files

This runs ruff (linting and formatting) and pytest.

Getting Help

For help on any command, use the --help flag:

scanpy-cli --help
scanpy-cli pp --help
scanpy-cli tl umap --help