Bump kallistobustools to 0.28.2

[gennadyFauna]

Changes:

• Bumped kb version to 0.28.2.
• Added support for the new nac workflow.
• Updated main.nf for ref and count to match nomenclature in meta.yml files: intron instead of introns
• Updated count to match ref and take in a workflow_mode variable. This fixes a bug: the results of the, e.g., lamanno pipeline are incorrect if the workflow is not specified; kb generates one matrix instead of two and ignores the intronic content.
• Updated tests to account for new quantification logic, removing the nucleus option and adding the nac one, and passing the workflow variable directly into count.
• Updated the test outputs.

PR checklist

Closes #XXX

• 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

[gennadyFauna]

Note that is a placeholder until the bioconda-recipes PR bioconda/bioconda-recipes#45846 is resolved. Due to incompatibilities between kallisto|bustools and other versions of kallisto, the relevant Docker container is broken.

[gennadyFauna]

The upstream conda distributable has been rebuilt on the relevant repositories; the Docker container (etc.) is now functional.

[CarsonJM]

Great work! At the moment we are trying to migrate modules from pytest to nf-test (https://nf-co.re/docs/contributing/modules#migrating-from-pytest-to-nf-test) would you be able to follow this guide and add nf-test?

[gennadyFauna]

I appreciate it! Unfortunately, at this time, I do not think I am comfortable enough with the pipeline to convert the pytest checks to nf-test, and would prefer to handle that in a future PR.

[gennadyFauna]

Great work! At the moment we are trying to migrate modules from pytest to nf-test (https://nf-co.re/docs/contributing/modules#migrating-from-pytest-to-nf-test) would you be able to follow this guide and add nf-test?

I think there is a test on the dev branch nf-core/scrnaseq written by @fmalmeida that could be straightforward to repurpose. I have already adapted it to the new workflow in the scrnaseq module. But I am not sure how I would insert it here.

[CarsonJM]

@gennadyFauna I went through and added the nf-tests if you are interested in taking a look!

[CarsonJM]

Here is a brief summary of how I went about this:

1. Ran the following code to add the appropriate files:
   nf-core modules create kallistobustools/ref --migrate-pytest
   ``nf-core modules create kallistobustools/count --migrate-pytest`
2. I added the files that were used as input in the pytest to the nf-test
3. I ran the nf-tests using the following command to see if there were any inconsistent hashes in the output
   nf-core modules test kallistobustools/ref
nf-core modules test kallistobustools/count
4. There were inconsistent hashes (when the tests fail is will show you which files had changed md5sums). So I added a line to test all files except those with inconsistent hashes:
   path(process.out.count.get(0).get(1)).list().findAll { file(it.toString()).name != "run_info.json" && file(it.toString()).name != "kb_info.json" },
5. I separate tests for the inconsistent files where I read the lines that don't contain dates/times/runtimes (these are typically what cause inconsistent hashes)
   file(path(process.out.count.get(0).get(1)).list().find { file(it.toString()).name == "kb_info.json" }.toString()).readLines()[2..10],
6. After going through this process for all inconsistent files and both modules, the tests passed!

Totally okay if this doesn't make sense yet, I just wanted to explain the changes I made and hopefully provide an intro to how nf-tests work!

[jasmezz]

Great, another nf-test migration 🎉 Check my comments and update if you find necessary.

[jasmezz]

Cannot check for file content for these index files?

[CarsonJM]

I think we probably could, but I honestly didn't look into it too much haha. I think you're right so I'll see what I can do before merging

[jasmezz]

Would make it slightly more robust. Checking for existence is my last resort when md5 or file content checks are impossible (e.g. a binary file with changing content due to timestamp or something).

[CarsonJM]

@gennadyFauna In short, the changes that @jasmezz made to the stub section were to remove unnecessary code. The stub section is basically to perform a "dry run" of a module, so only the outputs are created via touch but nothing is actually run.

The great benefit of this is that you can very quickly make sure that your inputs/outputs chain together correctly without actually running all of the tools.

[CarsonJM]

@jasmezz The index is in some compressed format and I couldn't quickly find out which. Merging for now get this in for @gennadyFauna, but definitely something to look into in the future

[gennadyFauna]

@jasmezz Thank you – I see, this is for the tests, then. I will read the docs in more detail.

@CarsonJM that should be the kallisto index. But I'm not sure there is documentation about its specifics.