Genomics · Database
CZ CELLxGENE
A single-cell platform to explore, download, and analyse standardised human and mouse transcriptomic atlases.
What CZ CELLxGENE is
CZ CELLxGENE Discover is a genomics platform developed by Chan Zuckerberg CELL by GENE to facilitate the study of human and mouse tissues at cellular resolution. It provides access to standardised single-cell expression atlases covering healthy states and diseases.
The resource addresses the difficulty of finding and analysing disparate single-cell transcriptomics data by gathering hundreds of collections and datasets into a unified corpus. Researchers can query normalized gene expression across tens of millions of cells, inspect cell lineage relationships, and download curated subsets for downstream investigation.
How it is organised
Data within the corpus is organised into standardised collections and datasets representing healthy and diseased tissue biology. Users can query and extract custom subsets of this standardised data directly into R and Python using the Census application programming interface (API).
For interactive exploration, the platform provides a no-code visual interface called Explorer to analyse expression against spatial, environmental, and genetic variables. Researchers can run differential expression tests between custom groups of cells based on metadata, or use the Annotate tool to assign and refine cell type labels directly on datasets.
Key facts
- Total records
- 100M+ cells
- Unique cells
- 33M+
- Datasets
- 436 datasets (with 1,000+ across collections)
- Cell types represented
- 2.7K+ cell types
- Cell types in Cell Guide
- 700+
- Organisms covered
- Human and mouse
- Supported programmatic interfaces
- Census API in Python and R
What you can use it for
- Differential expression testing
- Select two custom cohorts of cells based on metadata annotations to determine top differentially expressed genes across the corpus.
- Interactive visual analysis
- Inspect individual datasets in a no-code browser interface to examine how gene expression patterns vary with genetic, spatial, and environmental factors.
- Programmatic data slicing
- Extract custom slices of standardised single-cell data directly into Python or R workflows using the Census API.
- Cell type lookup and annotation
- Explore marker genes, lineages, and definitions for over 700 cell types in the Cell Guide, or assign novel annotations to datasets using CZ CELLxGENE Annotate.
Query it here
The research agent queries CZ CELLxGENE directly — no account, no API key — and every answer links back to the record it came from. Ask in plain language and it writes the query.
Strengths and limits
Good at
- Hosts a standardised corpus containing tens of millions of single-cell transcriptomic profiles.
- Provides programmatic integration with Python and R via the Census API.
- Includes a no-code visual Explorer for identifying cell subtypes and analysing gene expression.
- Incorporates a dedicated Cell Guide detailing marker genes, definitions, and lineages across hundreds of cell types.
Questions
Can I search CZ CELLxGENE without writing code?
Yes. The research agent queries CZ CELLxGENE directly, with no account or API key, and cites the record behind every answer. Ask in plain language — for a gene, a protein, a compound or a study — and it builds the query for you.
Can I reuse data from CZ CELLxGENE?
Portal CC BY 4.0; each dataset carries its collection's licence. Terms: https://cellxgene.cziscience.com/docs/08__Terms%20of%20Use
Which programming languages are supported by the Census API?
The Census API provides direct programmatic access to custom slices of standardised cell data in Python and R.
What organisms are covered in the database?
The platform contains datasets characterizing both human tissues (in health and disease) and healthy mouse tissues.
What information does the Cell Guide provide?
The Cell Guide serves as an encyclopedia for over 700 cell types, detailing lineage relationships, marker genes, definitions, and relevant datasets.
Written from CZ CELLxGENE's own website and reference material. Statements citing a figure that does not appear there are removed before publication. Licence and running details come from our registry.