search

genome / docker-immuno_tools-cwl

0 stars 0 forks source link

Optimize OptitypePipeline.py input #6

Open johnegarza opened 5 years ago

johnegarza commented 5 years ago

https://github.com/genome/docker-immuno_tools-cwl/blob/ef53b1e390a52280b362cd8a62237cd210ff4b71/optitype_script.sh#L55

Currently, our script provides 2 fastq files to Optitype as input, which results in an unnecessary second alignment by Optitype itself as part of its internal pipeline. Providing mapped and filtered bams as input skips this step; this should also allow us to avoid a few other conversion steps in our prep script.

johnegarza commented 5 years ago

Update- providing our own bams as input may not work; if this is the case, it might not be possible to use a custom HLA reference (optitype has one built in) without modifying the optitype code itself, but I need to verify this