CirroCirrospatial data studio

Appearance

Public Data ​

Open the web app (Open the app at the top of this site). It opens straight onto Spatial Data Studio's public sections, with no account. Public data is read-only: you can look and run every analysis, and what you save stays in this browser (Projects). Log In (Cirro) at the top right adds your own data (Your Data (Cirro)).

Finding a dataset and looking at it work the same for both: Find a dataset and Look at a section cover them. This page covers what public data is, which of it your browser lists, and how to show your own.

Public projects ​

Public data comes in projects, as all data in the app does. A public project is a project anyone can open with no account: a Cirro publication, copies of Cirro datasets in a public bucket, the same for every visitor. The app opens on Spatial Data Studio's own, Spatial Data Studio examples, where a section appears once it is published there. Its sections are read straight from where they are published.

Choose which public projects to list ​

Settings (in the top bar, or in your account menu once logged in), or Manage public projects… at the bottom of the project list, lists the public projects this browser shows. Each one gives its name, its description and the address it is read from, and says whether it is Spatial Data Studio's own or one you added, which this app hasn't reviewed.

  • Add a public project by pasting its address: a Cirro publication's root (the https://…/published/<name> address its publisher gives), or a folder with an index.json (Show your own data). The app checks that one of the two answers there before listing it. Other Cirro apps publish their public data the same way, so one of theirs can be added here; the app shows what it can read in it.
  • Remove drops one from the list; nothing is deleted. Spatial Data Studio's own can be removed too, and Bring back Spatial Data Studio's public project restores it.

The list is kept in this browser only: another browser, or a private window, starts with Spatial Data Studio's own.

Show your own data ​

A public project needs no Cirro: any folder of SpatialData stores, zipped (.sdata.zarr.zip), with an index.json at its root listing them, is one. Put it anywhere a browser can read files from another site with byte ranges (CORS), such as an S3 bucket set up for it, and add the folder's address in Settings. Each entry names a store and, if you like, how the map shows it:

json
{ "title": "Our lung sections",
  "description": "Xenium sections from the lab, raw and clustered",
  "checkpoints": [
    { "path": "raw/lung.sdata.zarr.zip", "label": "Lung (raw counts)", "folder": "Xenium" },
    { "path": "clustered/lung.sdata.zarr.zip", "label": "Lung, clustered", "folder": "Xenium",
      "from": ["raw/lung.sdata.zarr.zip"], "tags": ["tumor"],
      "credit": { "source": "…", "url": "…", "by": "…", "license": "CC BY 4.0" } } ] }

label and description name and describe it, folder files it (a / nests folders), from draws its lineage from the stores it was made from, and credit says whose data it is. Anyone who adds the address sees the same project, and nothing about it passes through Spatial Data Studio. To show one store on its own, with no index.json and nothing to add in Settings, link to it instead. A pipeline run's results/ folder already carries an index.json of this kind. CHECKPOINT_FORMAT.md lists every field.

Whose data it is ​

A public dataset says where it comes from. The line above an open section names the dataset it was made from, linked to its page, with its license, and Details says what was done to the data to make the section. A publication carries this in its datasets' credit.<field>: <value> tags, an index.json in each entry's credit. The map files each dataset in its folders.