Add singlem

[dgodin19]

Summary

This PR is an initial attempt to add support for SingleM to the profiling layer of the pipeline.

This work was started during the March 2026 nf-core hackathon by Dennis and Jack.

Changes

The following changes were made:

• Added a singlem module to modules/nfcore/singlem/singlem
• Integrated SingleM into the profiling.nf subworkflow
• Updated nextflow.config
• Added example samplesheet.csv and databases.csv files for testing

Running the module

I was able to run the module successfully with:
nextflow run main.nf -profile conda --input samplesheet.csv --databases databases.csv --outdir test_out

Database setup

SingleM requires pre-downloaded marker gene data. This must currently be done manually before running the pipeline.

Example workflow:

1. Create a conda environment with SingleM

2. Run:
   singlem data --output-directory /path/to/dbs

3. Add the resulting directory to the databases.csv

Example database and sample sheets have been included in the repository for reference.

Container issues

I attempted to run the module with both Docker and Singularity profiles. However, I encountered an error related to the -ps option in Nextflow. The exact error was something like:
Command error: Command 'ps' required by nextflow to collect task metrics cannot be found

The container referenced in the SingleM documentation appears to be minimal and may be read-only. Because of this, it may be preferable to build a custom container including:

• SingleM
• required dependencies
• writable data directory

and publish it to quay.io.

Remaining issues / TODO

The module produces output, but there are still some integration steps remaining:

• Integration with taxpasta (currently taxpasta does not support SingleM output format)
• Adding SingleM results to MultiQC

I started reviewing taxpasta documentation but did not complete this integration.

Notes

This PR is intended as a starting point for discussion and further development.

[jfy133]

@dgodin19 confirmed on Slack this was a hackathon project, and likely would not be able to continue

However this is a good scaffold we can continue from (thank you for kicking it off @dgodin19 !), so we leave the
PR open to update in the future.

Some initial notes from a quick review:

1. The module is currently actually a local module but in the wrong directory. We need to convert that to a proper nf-core module, following the nf-core standards (E.g. use of biocontainers etc)
2. Some files were incorrectly edited/added (.gitattributes, databases_empty.csv)

But the profiling code looks good, for example, and the new parameters in nextflow.config.