Have a list of default genomes that users can start with

[cmdcolin]

If there was a small list of genome such as human, mouse, arabadopsis, etc to kickstart JBrowse/JBrowse desktop with it might lower barrier to entry

[rbuels]

Yes we need this. Where do you think we might serve these from? Perhaps an S3 bucket might be good.

[garrettjstevens]

Totally agree. It's easy to find data, but hard to find it hosted along with an index. If we could set up a public S3 bucket with a few readily accessible genomes it would be awesome. If we could even host a GFF3 of the genes for that genome so they could have a sample track to look at, too, it might be nice.

[cmdcolin]

This bug will also need solving https://github.com/GMOD/jbrowse/issues/1269

[scottcain]

For the big NIH MODs, there is some data already in public S3 buckets (refSeqs, genes and UCSC conservation tracks), and more will be coming. That's for yeast, fruitfly, worm, zebrafish, mouse, rat and human (not a MOD, but you get the idea). Once there is a build with those included, it would be easy to go to other MODs and say "turn on CORS and establish a standard directory structure that doesn't change from release to release, and we'll include yours too". For the aforementioned species, we can also probably get at least some of them to turn on CORS and pull data directly from them. Only some though, I've had conversations with a few of them and they indicated that IT won't let them turn on CORS.

[cmdcolin]

JBrowse desktop actually does not need CORS

[cmdcolin]

It uses node-fetch instead of the browser fetch just for the reasoning there