---
id: cellxgene-census
version: "1.18.0"
license: MIT
license_treatment: permissive
maintenance: active
---

# cellxgene-census — API to facilitate the use of the CZ CELLxGENE Discover Census. For more information about the API and the project visit https://github.com/chanzuckerberg/cellxgene-census/

License: permissive · Maintenance: active · Downloads: 76.9K/mo

## What it is and what it does

cellxgene_census is a client library for accessing the CZ CELLxGENE Discover Census, a curated repository of single-cell genomics data. It abstracts the complexity of querying and retrieving cell metadata and gene expression measurements from a large distributed dataset, allowing researchers to filter cells by biological attributes (tissue, cell type, disease status, etc.) and fetch results as pandas DataFrames or other standard formats.

The package depends on tiledbsoma for efficient sparse array access, anndata for in-memory representation of single-cell data, and s3fs for cloud storage integration. It is designed for bioinformaticians and computational biologists who need programmatic access to reference single-cell datasets without managing raw files locally.

Use it for:

- Query cell metadata across tissues and organisms to identify cells matching specific biological criteria for downstream analysis
- Retrieve gene expression matrices for a subset of cells and genes to train machine learning models on reference data
- Integrate Census data with local analysis pipelines by fetching pre-filtered cell populations as DataFrames
- Explore cell type distributions and tissue composition across the human and mouse reference atlases

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Provides a Python API to query and retrieve cell metadata and expression data from the CZ CELLxGENE Discover Census, a large-scale single-cell genomics dataset.

Yes, if you work with single-cell genomics and need programmatic access to the CELLxGENE Census. The package is actively maintained, has no known vulnerabilities, uses a permissive MIT license, and installs with low friction. It is appropriate for research and production use. Install only if you have network access to the Census service and require Python 3.10 or later.

## Install

pip install cellxgene-census
uv add cellxgene-census
poetry add cellxgene-census

## Installing cellxgene-census

Before you install: Low friction installation with a pure-Python wheel. Actively maintained as of 2026-08-04 with recent releases. Requires Python 3.10 or later.

License in practice: MIT license permits commercial and private use with minimal restrictions—suitable for most research and production contexts.

Quickstart:

pip install cellxgene_census

import cellxgene_census

with cellxgene_census.open_soma() as census:
    cell_metadata = cellxgene_census.get_obs(
        census,
        "homo_sapiens",
        value_filter="cell_type == 'neuron'",
        column_names=["cell_type", "tissue"]
    )

Requires Python 3.10 or later and network connectivity to the CZ CELLxGENE Discover Census service.

Verify before relying:

- Whether the package supports offline access or requires live network connectivity to the Census service
- Performance characteristics when querying large cell populations or complex filters
- Data versioning and reproducibility guarantees across different Census releases

## Package facts

- License: MIT (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: active
- Downloads: 76.9K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

single-cell genomics data access, cellxgene census api, cell metadata query, gene expression data retrieval, bioinformatics data portal, single-cell rna-seq database, single-cell-genomics, bioinformatics, data-portal

[View on SkillFed](https://skillfed.io/packages/cellxgene-census) · [View on PyPI](https://pypi.org/project/cellxgene-census/)