Open ajo2995 opened 7 years ago
We don't have a CRAMtoBAM service, but it should be straightforward to make one. As for CRAM index files, it does make grabbing the same data much quicker. However you can run into collisions (2 cram files with the same name), so if this will be public or used by a larger group you may want to automatically delete the indices every so often (e.g once a day/week).
One thing to be potentially note is that a CRAMtoBAM minion service will result in a stream of BAM data. My guess is that biodalliance is expecting a bam url, so to hook this up my require some clever fiddling with biodalliance.
I've been using the biodalliance embedded genome browser, but it doesn't support CRAM formatted files, so I was thinking creating a minion web service to wrap a samtools conversion command. Has anyone already done this? If not, do you have any deployment suggestions such as caching the CRAM index files or "BAM" content for a given range query?