Hardcoded igv database

[marimeireles]

We have the igv database hardcoded in the code as a json. I wonder how often it changes and if this should be dynamic and fetch it from somewhere? Or if we should leave like it is.

PUBLIC_GENOMES_FILE = 'public_genomes.json'

[harrouet]

public_genomes.json was taken from the public URL https://s3.amazonaws.com/igv.org.genomes/genomes.json, as indicated at the end of this documentation page. I am not sure if it changes often, but I thought that it was not meant to be loaded each time a new notebook starts. Ideally it would be better to have a cached version in the package, and check for new versions once in a while, e.g. if instantiating an unknown genome.

[marimeireles]

Yeah! Seems like a good idea. Thanks for the input @harrouet. I'm mainly creating these issues as a form to keep track on things I'm thinking. I'm giving a review in ipyigv.

[harrouet]

Sure, feel free to reach out in case of question!

[marimeireles]

Thanks :)