alterlab-cellxgene

Query CELLxGENE Census data for single-cell datasets and expression matrices.

58|9|Updated Mar 16, 2026
One-click install
npx skills add https://github.com/AlterLab-IEU/AlterLab-Academic-Skills --skill alterlab-cellxgene
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: alterlab-cellxgene
Source: https://github.com/AlterLab-IEU/AlterLab-Academic-Skills/tree/main/skills/bioinformatics/alterlab-cellxgene
Command: npx skills add https://github.com/AlterLab-IEU/AlterLab-Academic-Skills --skill alterlab-cellxgene

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Programmatic access to the CZ CELLxGENE Census enables researchers to discover, filter, and analyze millions of single-cell records across thousands of datasets without manual data wrangling.

Core Features & Use Cases

  • Metadata exploration: Retrieve cell and dataset metadata with sparse queries to understand available data.
  • Expression querying: Load small-to-large expression matrices via get_anndata, axis_query, and related helpers for downstream analyses in Python.
  • Machine learning and integration: Train models with PyTorch, and integrate results with Scanpy for end-to-end workflows; multi-dataset integration supports cross-tissue analyses.
  • Two-step exploration and query: First explore metadata, then execute targeted queries to retrieve focused cohorts or genes.

Quick Start

Open a census context and execute a basic query to list B cells in lung tissue where is_primary_data is True.

Frequently Asked Questions about alterlab-cellxgene

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

FAQPage Schema
How do I query CELLxGENE Census data to filter single-cell datasets by tissue and cell type?

The CELLxGENE Census provides programmatic access to millions of single-cell records across thousands of datasets, enabling you to discover, filter, and analyze data without manual data wrangling.

Can I retrieve expression matrices from CELLxGENE Census for downstream analysis in Python?

Yes, CELLxGENE Census integrates with Scanpy for end-to-end workflows and supports training models with PyTorch, enabling multi-dataset integration for cross-tissue single-cell analyses.

Does CELLxGENE Census work with PyTorch and Scanpy for machine learning workflows?

You can retrieve expression matrices from CELLxGENE Census using get_anndata or axis_query helpers, which return AnnData objects suitable for immediate downstream analysis in Python.

What is the best way to explore metadata before executing a targeted query in CELLxGENE Census?

The best approach is a two-step process: first explore cell and dataset metadata with sparse queries to understand available data, then execute targeted queries to retrieve focused cohorts or genes.

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

Yes, accessing CELLxGENE Census requires TileDB-SOMA-based APIs and the cellxgene_census Python package to query large-scale single-cell datasets and retrieve standardized metadata.