samtools/merge: emit .bai/.crai index files by default (fixes #4261)

[ramakrishna-p21]

PR checklist

Closes #4261

• 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.
• Broadcast software version numbers to topic: versions - See version_topics
• 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 samtools/merge --profile docker

Description of Changes

This PR updates samtools/merge to automatically generate and emit index files (.bai for BAM and .crai for CRAM) alongside the merged alignment file, resolving #4261.

Key Modifications

• main.nf: Added --write-index and ##idx## syntax to the samtools merge command execution to emit .bai for BAM files and .crai for CRAM files by default. Updated the stub: block to touch the corresponding index file.
• meta.yml: Updated documentation for the emitted index channel pattern.
• tests/main.nf.test: Removed manual --write-index overrides from module test configs.
• tests/main.nf.test.snap: Refreshed snapshots to reflect the generated .bai and .crai indices.

---

Note: Developed and verified with assistance from Antigravity. All test fixtures, linting, and biological logic were manually reviewed and validated against nf-test.

[SPPearce]

What happens if the bam/cram files are not in coordinate sort order?

[SPPearce]

Now I could be convinced to have an input channel to make the indexes, rather than the horrific --write-index mess.

[ramakrishna-p21]

Thanks for the review, @SPPearce. That's a great point about the sorting edge cases, and I agree on dropping the ##idx## syntax.

If we handle this through the input channel, would adding an optional boolean (like val(create_index)) to trigger a separate samtools index step work best for you?

Let me know how you'd prefer the input signature structured so we don't break existing pipelines, and I'll get the PR updated!

[SPPearce]

Hi @ramakrishna-p21 , can you take a look at how samtools/sort implements this, with the input channel there.
I've had approval from the other maintainers to have this input channel like this.

[SPPearce]

Also please join the community via the slack channel github-invitations.

[ramakrishna-p21]

Hey @SPPearce, I've refactored the module to use the index_format input channel, matching the samtools/sort architecture exactly. I also updated the downstream bam_applybqsr subworkflow to dynamically pass crai or bai based on the output format, and regenerated all tests and snapshots. Please take a look at the updated changes when you have a moment. Also, I've just submitted my GitHub invitation request in the slack channel as suggested!

[ramakrishna-p21]

Hi @SPPearce, I've updated everything based on your feedback:

• Swapped bai to csi in bam_applybqsr and re-recorded the snapshot

• Added the sorting note to meta.yml

• Cleaned up main.nf.test with sanitizeOutput(readsMD5Keys: ...) and refreshed the snapshots

Local tests and linting are all green. Let me know if you spot anything else!

[SPPearce]

Thanks!
Minor tweak to use the test data param.

[SPPearce]

Huh, looks like that didn't work. So just put it back to how it was

[ramakrishna-p21]

Done, put it back to how it was! Tests and linting are all green again. Feel free to merge whenever you're ready!