cellxgene-census

Query CELLxGENE Census single-cell data via TileDB-SOMA and Python APIs.

1|Updated Mar 11, 2026
One-click install
npx skills add https://github.com/SciMate-AI/scicli --skill cellxgene-census-scimate-ai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: cellxgene-census
Source: https://github.com/SciMate-AI/scicli/tree/main/internal/skills/bundled/claude-scientific-skills/skills/cellxgene-census
Command: npx skills add https://github.com/SciMate-AI/scicli --skill cellxgene-census-scimate-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Programmatic access to the CELLxGENE Census enables researchers to query large-scale single-cell datasets across tissues, diseases, and cell types, returning metadata, counts, and expression data for downstream analyses.

Core Features & Use Cases

  • Metadata exploration: browse datasets and cell annotations without loading heavy matrices.
  • Expression queries: retrieve Anndata-compatible data or raw counts and integrate with Scanpy or PyTorch.
  • Cross-dataset analyses: perform multi-tissue or disease-focused queries with optional out-of-core processing.

Quick Start

Open the Census and filter for primary Homo sapiens data to retrieve T cells from lung tissue.

Frequently Asked Questions about cellxgene-census

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

FAQPage Schema
How do I query single-cell gene expression data from CELLxGENE Census?

Query single-cell gene expression data by opening the Census and filtering for primary data across tissues and cell types using axis_query and Python APIs to retrieve Anndata-compatible outputs.

Can I retrieve metadata from scRNA-seq datasets without loading the full expression matrix?

Yes, scRNA-seq metadata exploration allows browsing datasets and cell annotations directly via TileDB-SOMA obs/var structures without loading heavy matrices into memory.

Does CELLxGENE Census work with Scanpy and PyTorch for downstream analysis?

CELLxGENE Census integrates with Scanpy and PyTorch by retrieving Anndata-compatible data or raw counts for downstream analytics and model development workflows.

What is the best way to perform cross-tissue single-cell queries on large-scale datasets?

Cross-tissue single-cell queries use axis_query to perform multi-tissue or disease-focused filtering, supporting both in-memory and out-of-core processing for large-scale datasets.

Do I need TileDB-SOMA to access CELLxGENE Census data programmatically?

Yes, the programmatic access relies on TileDB-SOMA structures including presence matrices and obs/var arrays to query and manipulate single-cell data efficiently.

How do I filter primary Homo sapiens data to retrieve specific cell types from lung tissue?

Open the Census and apply dataset-level filtering using axis_query to select primary Homo sapiens data, retrieving specific cell types from lung tissue via Python APIs.