enable a pre-built bowtie2 index for rrna removal

[emmanuel-tan]

Summary

Adds support for supplying a pre-built Bowtie2 index for rRNA removal (--ribo_removal_tool bowtie2), so users don't have to rebuild the index from --ribo_database_manifest on every run.

Addresses #1731

Changes made

• Added --bowtie2_rrna_index to nextflow_schema.json (path or .tar.gz archive) with description/help text.

• subworkflows/local/prepare_genome_indices/main.nf:

  • Added bowtie2_rrna_index as a take: input and a new emit output.
  • Added a bowtie2_rrna' branch to prepare_tool_indices (only added when ribo_removal_tool == 'bowtie2' and an index path is supplied).
  • New step 3b) untars the archive via a new UNTAR_BOWTIE2_RRNA_INDEX module (only if the path ends in .tar.gz) or passes the directory straight through if it's already unpacked. No index is built here as that stays in fastq_remove_rrna, this subworkflow only resolves/untars a pre-built one.

• workflows/rnaseq/main.nf: make_bowtie2_index is now false whenever --bowtie2_rrna_index is set, skipping on-the-fly index construction.

• main.nf: params.bowtie2_rrna_index is threaded into PREPARE_GENOME_INDICES, and its resolved bowtie2_rrna_index output is passed to RNASEQ.

• subworkflows/nf-core/fastq_remove_rrna/main.nf: reverted to the nf-core version of this workflow; an earlier commit of mine incorrectly modified this.

• tests/remove_ribo_rna.nf.test (+ .snap): two new nf-test cases (--bowtie2_rrna_index, normal and -stub) asserting the index-build steps (SEQKIT_REPLACE*, BOWTIE2_BUILD) are skipped and UNTAR_BOWTIE2_RRNA_INDEX runs instead. Pending PR that adds a small test index at add small bowtie2 index for tests test-datasets#2174 .

How index resolution now works

1. --bowtie2_rrna_index unset → make_bowtie2_index = true, and fastq_remove_rrna builds the index from --ribo_database_manifest FASTAs as before.
2. --bowtie2_rrna_index set → prepare_genome_indices resolves it (untarring if needed) and that channel is passed straight through to fastq_remove_rrna, which skips the build step entirely.

Outstanding items

• The new tests currently do not point to a bowtie2_rrna_index parameter. They were tested locally pointing to a test index at a personal fork. Need to point to nf-core/test-datasets before merge (add small bowtie2 index for tests test-datasets#2174) and regenerate snapshots.
• The nf-core subworkflow FASTQ_QC_TRIM_FILTER_SETSTRANDEDNESS currently initialises ch_bowtie2_index to channel.empty(), preventing any provided index from being utilised for the FASTQ_REMOVE_RRNA step. Separate PR here remove setting of bowtie2_index to an empty channel modules#12490

PR checklist

• This comment contains a description of changes (with reason).
• If you've fixed a bug or added code that should be tested, add tests! (nf-test cases added; blocked on official test-datasets index — see above)
• If you've added a new tool - have you followed the pipeline conventions in the contribution docs
• If necessary, also make a PR on the nf-core/rnaseq branch on the nf-core/test-datasets repository.
  • add small bowtie2 index for tests test-datasets#2174
• Make sure your code lints (nf-core pipelines lint).
• Ensure the test suite passes (nextflow run . -profile test,docker --outdir <OUTDIR>).
• Check for unexpected warnings in debug mode (nextflow run . -profile debug,test,docker --outdir <OUTDIR>).
• Usage Documentation in docs/usage.md is updated.
• Output Documentation in docs/output.md is updated.
• CHANGELOG.md is updated.
• README.md is updated (including new tool citations and authors/contributors).

[emmanuel-tan]

Hi @pinin4fjords ! I just wanted to highlight this PR for your review. It does depend on PRs in the nf-core/test-datasets and the nf-core/modules repos so leaving as a draft first. Hope the approach taken makes sense! Open for further thoughts and discussion.