Viewing and Loading#
This section answers: “How do I look at my object from R?” and “I wrote a store
with cellucid_prepare() — how do I view it, and how do I share it?”
cellucid_view(object) opens a Seurat, SingleCellExperiment or
SummarizedExperiment object (or any file the web app reads) in the RStudio or
Positron Viewer pane, a browser, or – in a Jupyter notebook – the cell’s
output, with no file written and no Python; cellucid_serve() serves it to a
browser elsewhere.
cellucid_prepare() writes a Cellucid-ready AnnData: a standard AnnData Zarr
store, either a .zarr folder (format = "zarr") or one .zarr.zip file (format = "zarr.zip"). The
web app opens it like any other AnnData store, so there is no separate format to
load.
On your own computer: write a
.zarr.zipand open it with Open… › File….Shared with others: put the
.zarrfolder or the.zarr.zipfile on a web host (GitHub, Hugging Face, Zenodo, a bucket) and open it with the Link or server bar.
Note
The web app’s own loading documentation is in Web App Guide; every way to open data from a web address is in Open from a link.
Open a .zarr.zip from your computer, or a store from a web address.
Publish a store on GitHub, Hugging Face, Zenodo or any web host, and share one link.
Read the store back in R and in Python’s anndata, and diagnose loading failures.
cellucid_view(seu): the object in the Viewer pane, from memory.
cellucid_view(seu) in IRkernel: the viewer in the cell’s output, no port.
cellucid_serve(), SSH tunnels, RStudio Server and Workbench proxies.
Commands from R, events back, a selection or the session in R.
RNA + ADT or ATAC as modalities of a Cellucid-ready MuData.