Mark Keller

How to incrementally add this: - Data loader classes can optionally include a `loadArrow()` method. The return value of this function will be used to insert into the database. -...

```js // Pseudocode useInsertedData( [ ['obsEmbedding', datasetId, matchOn], ['featureSelection', datasetId, matchOn, featureSelection], ['obsSets', datasetId, matchOn, obsSetSelection], ['sampleEdges', datsestId, matchOn] ] ); // Internally: multiple useQueries that perform loader.loadArrow() followed by...

https://loaders.gl/docs/arrowjs/developer-guide/converting-data

https://spatialdata.scverse.org/en/latest/tutorials/notebooks/datasets/README.html

Depends on https://github.com/vitessce/vitessce/pull/1728

We will also want to embed an example notebook in the documentation that demonstrates how to use the implementation. These notebooks are located at https://github.com/vitessce/vitessce-python/tree/main/docs/notebooks

Once this is complete, we should also deploy an example and/or a tutorial.

TODO: investigate kerchunk as an alternative solution

TODO: wrap https://github.com/jrobinso/hdf5-indexed-reader as a zarrita store

Related to https://github.com/vitessce/vitessceR/issues/55