look into loading gene expression data into the model

[vincerubinetti]

I'm told by @cgreene that this would take significant effort. It seems like there are some backend technical challenges as to why this data can't be put into the database at this time.

If the user were to have access to the gene expression data, it would pretty simple to add functionality where they can drag and drop the data into the browser, and the samples table would then include the gene expression data (perhaps not all visible at once, but would be there when downloading the table as .tsv).

Something to look into later, as it doesn't seem to be a crucial feature right now.

Per Casey, a potential solution:

if one wants to do it in a performant way, the way to do it is to shard by experiments and pre-cache that in something like an s3 bucket and point the user to the appropriate file

[cgreene]

My suggestion was to pre-shard the dataset into experiments and store those somewhere + then link to those sharded files. Seems better than selecting on the fly since all of the experiments are pre-known.