parabricks/applybqsr version updated and tests migrated to nf-test

[gitanoqevaporelmundoentero]

Tools applybqsr is updated to latest version (4.3.2-1) and tests were migrated to new convention (from pytest to nf-test).

PR checklist

Closes #6861

• This comment contains a description of changes (with reason).
• If you've fixed a bug or added code that should be tested, add tests!
• If you've added a new tool - have you followed the module conventions in the contribution docs
• If necessary, include test data in your PR.
• Remove all TODO statements.
• Emit the versions.yml file.
• Follow the naming conventions.
• Follow the parameters requirements.
• Follow the input/output options guidelines.
• Add a resource label
• Use BioConda and BioContainers if possible to fulfil software requirements.
• Ensure that the test works with either Docker / Singularity. Conda CI tests can be quite flaky:
  • For modules:
    • nf-core modules test <MODULE> --profile docker
    • nf-core modules test <MODULE> --profile singularity
    • nf-core modules test <MODULE> --profile conda
  • For subworkflows:
    • nf-core subworkflows test <SUBWORKFLOW> --profile docker
    • nf-core subworkflows test <SUBWORKFLOW> --profile singularity
    • nf-core subworkflows test <SUBWORKFLOW> --profile conda

[famosab]

Looks good so far.

[famosab]

We need to change the way the GPUs are accessed in the main.nf file. Maybe we can do that here aswell.

so the line
--num-gpus $task.accelerator.request \\

needs to be
$num_gpus \\

and add this above the script block
def num_gpus = task.accelerator ? "--num-gpus $task.accelerator.request" : ''

---

And we need to add below the label expression on the top of the file
label 'process_gpu'

[sateeshperi]

honestly no idea since I didn't test using singularity. I just borrowed that config block from nf-core/scdownstream

[famosab]

Ok small update from me: Our cluster does not support docker, only singularity and for some reason just adding the --nv flag to the singularity options did not work for me.

[sateeshperi]

could you try what we had previously ?

[famosab]

could you try what we had previously ?

No, I think the AWS instance was closed. I am trying to run it on other stuff we have locally but each one of those has some limitations.

[mashehu]

restarted the test instance

[famosab]

Snaps are updated. I could not solve the weird fq2bammeth behaviour.
Seems like the test for applybqsr somehow does not pick up and the singularity one for fq2bam fails?

[mashehu]

singularity hiccup is fixed. the other tests shouldn't be picked up...

[famosab]

@sateeshperi Now the fq2bammeth test fails with
.command.sh: line 2: bwameth.py: command not found

any ideas?

[sateeshperi]

that means the container profile is not invoked properly....hmmmm

[famosab]

Maybe we can remove it from this PR and tackle it separately. I would like to merge this to have the GPU profile in the CI ;)

[famosab]

@sateeshperi if you are fine with that then give us the approval :D

[famosab]

singularity hiccup is fixed. the other tests shouldn't be picked up...

I think the issue is back, back again @mashehu

[mashehu]

There is a weird ci run, where the profile is empty. No idea how that can be. Will investigate tomorrow

[famosab]

@mashehu ping :)

[famosab]

Now only docker fq2bam is failing with

    >   [PB Error 2024-Nov-15 11:59:36][src/main.cu:997] CUDA_CHECK() failed with no CUDA-capable device is detected (100), exiting.
    >   For technical support visit https://docs.nvidia.com/clara/index.html#parabricks
    >   Exiting...

[famosab]

If we can fix this and merge it all the other parabricks modules can be updated accordingly and a bunch of PRs closed :)