HiSeqX is demultiplexing on Hasta

[ingkebil]

As a member of TEAPOT, I want to be able to use fastq files from HiSeq X flowcells on Hasta, so I can analyze or reanalyze a case.

Problem: HiSeqX demultiplexing is now running on Rasta.

Suggested solution:

Partly solved here: https://github.com/Clinical-Genomics/demultiplexing/pull/74

Changes need to be made to demultiplexing scripts/hiseqx/xcheckfornewrun.bash, scripts/hiseqx/xdemuxtiles.bash, scripts/hiseqx/xdemuxtiles.batch

• [x] Have IT install same version of bcl2fastq that is installed on Rasta on Hasta

• [x] scripts/hiseqx/xcheckfornewrun.bash

  • Don't source bashrc, instead activate correct env (stage or prod)
  • Change params to point to correct directories

• [x] scripts/hiseqx/xdemuxtiles.bash

  • Don't source bashrc, instead activate correct env (stage or prod)
  • Change params to point to correct directories
  • Remove copycomplete logic that waits for jobs to be copied to a node

• [x] scripts/hiseqx/xdemuxtiles.batch

  • Don't source bashrc, instead activate correct env (stage or prod)
  • Change params to point to correct directories
  • No more copying from input
  • Change params of bcl2fastq to match the ones suggested for NovaSeq (see Novaseq US)
  • Refactor copy of output to rename of output

• [ ] remove the copycomplete.txt trigger from deliver and possibly add another trigger?

• [ ] activate the prod env in checkfornewdemuxonhasta.sh

• [ ] Create test specification

• [ ] Validate

• [ ] Add crontab entry: ... copy crontab entry with xcheckfornewrun.bash from rasta ...

[emiliaol]

@ingkebil Is this still happening in the background for non-complicated FCs? Or is the NAS access denied problem going to interfere with this?

[ingkebil]

This is not happening anymore.