Data Access / API

Programmatic Access to the BrainStorm Atlas

Downloadable resources, interactive exploration, and a programmatic interface in preparation

The BrainStorm atlas is currently distributed as downloadable resources and through an interactive species explorer, both available today. A dedicated programmatic API is in preparation: when it launches, endpoints and usage examples will be documented here. Until then, this page is the reference for the access routes that are already live and for the data products you can work with.

Status: API in preparation

A dedicated programmatic API is being prepared. No endpoints are live yet, so we do not document endpoints here today. When the interface is available, this page will carry the schema, request examples, response formats and a quick-start guide. Meanwhile, every data product underlying the atlas can be reached through the access routes below.

Access Routes

Ways to Reach the Data Today

Before a dedicated API ships, the atlas is served through three complementary routes: bulk download, per-species exploration, and documentation of the file formats. These routes remain fully supported and are the simplest way to work with the data now.

Route What it gives you
Bulk Data Downloadable datasets for whole-atlas, cross-species and per-species analysis, including the integrated matrix and per-species matrices described below. datasets.php →
Species Explorer Interactive per-species exploration of the 100 species, with detail pages for individual species through the catalog. species.php →
Species Details Dedicated content pages for each individual species, surfaced from the explorer catalog. Species detail →
Browsing & Comparison Interactive single-cell browsing and comparative views for exploring the atlas before downloading. browser.php →
File Formats Documentation of the file formats, conventions and schemas used across the downloadable resources and the explorer. docs.php →

The datasets route is the closest analogue to programmatic bulk access today: it delivers the underlying matrices and metadata in documented, reusable formats (see documentation). The species explorer and its detail pages are the per-species, query-oriented analogue. In most cases a research workflow can already be assembled from these two together with the format reference.

Data Products

What the Data Access Surfaces

The same data products a future API would serve are the ones distributed through the routes above. They are described here so you know what to expect, whichever route you choose.

Integrative

Integrated Atlas Object

The cross-species integrated atlas, harmonized across the 100-species analysis layer of 492,121 cells and 11,578 orthologous genes.

Annotation

Full-Cell Metadata

Metadata for every cell across the 4,414,988-cell full matrix, including the 912,605 cells passing QC, with cluster and annotation fields.

Per-Species

Per-Species Matrices

Individual matrices that underpin the species explorer, aligned to the shared cross-species ortholog universe for comparison.

Evolutionary

Cross-Species Ortholog Table

The orthology mapping used to align the species, the foundation of the comparative and cell-type analyses.

100
Vertebrate Species
analysis layer
4,414,988
Cells Full Matrix
all libraries
492,121
Analysis-Layer Cells
11,578 orthologous genes
43
Atlas-Level Clusters
cross-species annotation

Raw and processed single-nucleus RNA-seq data from the newly profiled species are deposited with the Genome Sequence Archive at the National Genomics Data Center in Beijing, and the integrated atlas object, full-cell metadata, per-species matrices and cross-species ortholog table are deposited in a public repository with a commitment of at least 5 years of availability. The accession identifiers are being finalized and will be listed here, on datasets and in the publications alongside the manuscript.

Workflows

Assembling a Data Workflow Today

Until the programmatic API ships, a complete analysis can be assembled from the existing routes in a few steps. The pattern below maps directly onto the interface the future API is expected to provide.

  1. Choose a scale

    For whole-atlas or comparative work, start from the datasets page. For a single species, open the species explorer and pick from the 100-species catalog.

  2. Review the schema

    Check the documentation for the file formats and fields before writing any code, so your scripts match the delivered structure on the first pass.

  3. Explore before downloading

    Use the cell browser and the comparative view to form hypotheses, then pull the specific matrices and metadata you need from the datasets route.

  4. Map to the analysis layer

    Filter to the analysis layer of 492,121 cells across 11,578 orthologous genes, using the full-cell metadata and cross-species ortholog table, to reproduce the atlas-level comparisons.

Programmatic Access

What the API Will Add

The dedicated API is being prepared to make these capabilities available as stable, queryable endpoints. The concrete design, including the final list of operations, is still being finalized, so this is a description of intent rather than an endpoints reference.

Bulk retrieval

Getting the integrated matrix, per-species matrices, and the full-cell metadata in machine-readable form, mirroring what the datasets route delivers.

Gene and cell queries

Subsetting the same analysis layer of 492,121 cells and 11,578 orthologous genes by species, cell type or gene of interest, without downloading whole files.

Species lookup

Programmatic access to the same species catalog that the explorer surfaces, spanning the 41 fishes, 38 mammals, 10 birds, 8 reptiles and 3 amphibians of the analysis layer.

Documentation first

When endpoints are finalized, this page will provide the full reference, example requests and responses, and a quick-start guide.

Get Started With the Data

Explore the atlas interactively, download the matrices, or read the format documentation while the programmatic API is being finalized.

Download Data Species Explorer Documentation