Inconsistency of alignment subworkflows and aligner modules

[Kevin-Brockers]

Have you checked the docs?

• nf-core website: troubleshooting
• nf-core modules documentation

Description of the bug

Hi all,

I'm a bit puzzled. I planned to update a few things in the chipseq pipeline and stumbled across an inconsistency.
Before changing / adapting some of the rather central modules (bwa or bowtie2), I'm seeking for some advice:

Some of the alignment subworkflows use the BAM_SORT_STATS_SAMTOOLS subworkflow, which now expects a ch_fasta_fai input channel.
The subworkflows: fastq_align_bowtie2, fastq_align_bwa or fastq_align_chromap have already been adapted accordingly.

However, bowtie2 and bwa modules currently only take ch_fasta as inputs.
Example:
Subworkflow: https://github.com/nf-core/modules/blob/master/subworkflows/nf-core/fastq_align_bowtie2/main.nf

Module: https://github.com/nf-core/modules/blob/master/modules/nf-core/bowtie2/align/main.nf

I see two options:
a) adapt the inputs for bowtie2/build, bowtie2/align, bwa/index and bwa/mem (and likely also the other bwa modules).
b) just go ahead and patch things to match the pipeline needs.

If we agree on a) I wouldn't mind to update the modules and open a PR later.

All best,
Kevin

Command used and terminal output

Relevant files

No response

System information

No response

[matthdsm]

I'd go with option A, having an index along for the ride is beneficial in most cases.

Disclaimer, I'm the author of the changes mentioned above

[Kevin-Brockers]

Ok perfect!

I see / completely agree with the reasoning, just wanted to quickly double check before changing things. I will prepare a PR then.