cellxgene-census

Query CELLxGENE Census single-cell expression data via Python with versioned SOMA access.

783|65|Updated Feb 27, 2026
One-click install
npx skills add https://github.com/LeonChaoX/qinyan-academic-skills --skill cellxgene-census-leonchaox
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: cellxgene-census
Source: https://github.com/LeonChaoX/qinyan-academic-skills/tree/main/skills/05-%E7%94%9F%E7%89%A9%E4%BF%A1%E6%81%AF%E4%B8%8E%E5%9F%BA%E5%9B%A0%E7%BB%84%E5%AD%A6/cellxgene-census
Command: npx skills add https://github.com/LeonChaoX/qinyan-academic-skills --skill cellxgene-census-leonchaox

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires cellxgene-census, and includes references (resource) components.

What problem does it solve?

You need programmatic access to large, curated single-cell datasets so you can query expression and cell metadata without manually downloading and cleaning different atlas versions.

Core Features & Use Cases

  • Census querying at population scale: Retrieve expression matrices and structured metadata across tissues, diseases, and cell types from the CELLxGENE Census.
  • Reproducible versioned access: Open stable or a specific census release to keep analyses consistent over time.
  • Multiple workflow modes: Use get_anndata() for in-memory analyses and axis_query() for out-of-core, chunked processing.

Quick Start

Use the cellxgene-census skill to query human RNA expression for a specific cell type and tissue from a chosen Census version.

Frequently Asked Questions about cellxgene-census

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I query large-scale single-cell expression data without downloading entire datasets?

Querying large-scale single-cell expression data programmatically is done using the CELLxGENE Census to retrieve expression matrices and structured metadata across tissues and cell types without manual downloads.

Can I use Scanpy with cellxgene-census for in-memory single-cell analysis?

Yes, you can use Scanpy with cellxgene-census by utilizing the in-memory get_anndata() workflow to retrieve expression matrices formatted for standard single-cell analysis.

How do I get human RNA expression data for a specific cell type and tissue from the Census?

To get human RNA expression data for a specific tissue and cell type, open a chosen Census release and apply metadata filtering to retrieve the targeted expression matrix.

What is the best way to process single-cell atlas queries that exceed available memory?

For single-cell atlas queries exceeding memory limits, use the out-of-core axis_query() workflow to perform chunked processing on large expression matrices.

How do I ensure reproducible dataset creation for cross-disease cell model training?

Ensure reproducible dataset creation for cross-disease cell model training by opening a specific versioned Census release, keeping your atlas comparisons and metadata queries consistent over time.

Do I need the cellxgene-census Python package for out-of-core axis_query workflows?

Yes, the cellxgene-census Python package is required, as it provides the versioned SOMA access needed to support both out-of-core axis_query() and in-memory workflows.