Claude Skill MartBrowse skillsQuick linesLearn by videoTerminal guideWhat is a Skill?
 
← Back to list

CELLxGENE Census Data Querying

Guides Claude to query and analyze the 200M+ cell CZ CELLxGENE Census of public single-cell and spatial transcriptomics data without downloading full datasets.

Data & AnalyticsAdvanced★ 33,030⑂ 3,248AI score 8/10Last updated: Aug 9, 2026

What it does

  • Writes cellxgene_census Python code to pull cell metadata and expression slices filtered by cell type, tissue, disease, assay, donor and more.
  • Returns small/medium slices as AnnData, or switches to out-of-core axis_query() processing when the slice won't fit in memory.
  • Enforces best practices: pin census_version for reproducibility, always filter is_primary_data == True, select only needed columns, and check the presence matrix before gene-level analysis.
  • Covers spatial Census access, Scanpy hand-off, PyTorch training via TileDB-SOMA-ML, and multi-dataset integration with batch-effect handling.
  • Includes a troubleshooting checklist for oversized queries, memory errors, duplicate cells, missing genes, and version drift.

Who it's for

  • Bioinformatics researchers and grad students working with single-cell or spatial transcriptomics.
  • Anyone benchmarking their own data against public reference atlases.
  • ML engineers training cell-type classifiers on millions of cells.

Example uses

  1. "Show the cell type distribution in human lung" → metadata-only get_obs query summarized with value_counts().
  2. "Get CD4/CD8A expression in T cells from COVID-19 donors" → two-step explore-then-query flow loading only the needed genes.
  3. "Compare macrophages across lung, liver, and brain" → cross-tissue query followed by Scanpy rank_genes_groups.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/K-Dense-AI/scientific-agent-skills/HEAD/skills/cellxgene-census/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on Claude Skill Mart.
Copy the skills/cellxgene-census folder from the GitHub repo K-Dense-AI/scientific-agent-skills into my ~/.claude/skills/cellxgene-census/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/K-Dense-AI/scientific-agent-skills.git && mkdir -p ~/.claude/skills && cp -r scientific-agent-skills/skills/cellxgene-census ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

  1. Set up a Python 3.10–3.12 environment (3.13 is not supported yet).
  2. Clone the repo: git clone https://github.com/K-Dense-AI/scientific-agent-skills.git
  3. Copy the skill into place: mkdir -p ~/.claude/skills && cp -r scientific-agent-skills/skills/cellxgene-census ~/.claude/skills/
  4. Install dependencies: uv pip install "cellxgene-census==1.17.*" (add the [spatial] extra for spatial data, or tiledbsoma-ml for model training).
  5. Restart Claude Code and try a prompt like "Query human lung cell types from the CELLxGENE Census".
  6. No authentication is needed for public Census data, but always estimate cell counts before loading large slices to avoid memory blowups.
View source on GitHub ↗License: MIT