Is there a way to use the display in IGV tool so I don't have to download the genome separately?

usegalaxy.org support
1

It seems like Galaxy has both the genome index and the bam file - is there a way to send that information to IGV in one go instead of downloading the reference genome and building a custom IGV genome from that before using the display link?

2

Hi @bwlang !!

The genome might be in IGV already but maybe not installed into your local copy? By default not everything is pre-loaded. There is an import function in the menu functions where you can check. If not available, then yes, you’ll need to index the custom genome in both places if you want to load multiple files into the same view, and have the base-level details.

Then, if your datasets in Galaxy have the same database (dbkey), when you use the link out the transfer function will route without both indexing steps. That database can be native to the Galaxy server or a custom build you create in your Galaxy account. Same for IGV. The genomes need to be “exact” of course, e.g. same chromosome identifiers/lengths or an error can come up.

Adding a summary since it hasn’t been posted in a while.

How to link data between Galaxy and IGV

Single file display based on coordinates without base-level details

  1. Use the link to IGV under a dataset’s visualization display function
  2. Wait for the one-off indexing to complete, and you’ll be automatically redirected

Single or Multiple file display based on coordinates with base-level details

  1. Reference Genome is indexed in IGV (native or custom)
  2. Reference Genome is indexed in Galaxy (native or custom)
  3. Reference Genome is exactly the same assembly/build and given same database (dbkey) in both
  4. Dataset(s) assigned that database (dbkey)
  5. Use the link to IGV under a dataset’s visualization display function

Let me know if that is enough info or not! Jen

3

Hi @jennaj - nice to hear from you!

I’m working on a genome that is not present in any IGV genome list (neither local nor distributed by them). In this case it’s a special genome that’s a composite of human and malaria, (in the unlikely event that is relevant).

I can get things working nciely by creating a .genome file or manually downloading the reference and loading that index that - but I wondered if galaxy could just serve up a URL to an fai file to define a genome “on the fly” in some way (to make it easier for collaborators). Sounds like no - I need to get IGV prepared with the correct genome before trying to get galaxy to talk to it.

Maybe a good feature for a hackathon project!

brad

4

Yes, while it can work that auto way for a single file (without the fasta, just displays the coordinates/data in space), that doesn’t work well for multiple files since IGV doesn’t know to stack these into the same browser. If there is a way around it, perhaps some better synch between Galaxy and IGV’s “custom genome” functions, I’m sure others would be interested too, and a hack seems like the right place. Maybe next summer in Brno :beers: