Closed nschcolnicov closed 17 hours ago
@apeltzer Just to clarify, do you want us to convert the COMPUTE_GENE_SCORES module into a single module subworkflow? Or were you envisioning something else?
Honestly speaking it might even make sense to make this a two - step process:
Hi @apeltzer I assessed what it would I understand that the scope of this ticket is to create a subworkflow that could work for at least both differentialabundance and nanostring.
I assessed both pipelines and which tools they use to achieve this:
I think that attempting to create a subworkflow that would work for both of these pipeline wouldn't be ideal, as I would be mixing a list of tools, having half being run only in one pipeline and the other only being run for the other pipeline. This would also involve having unnecesary input channels that are only used in one of the two pipelines, with the matrix counts file being the only input they have in common.
Instead I moved the compute gene scores and heatmap plot modules into a subworkflow, which we can use in this pipeline: https://github.com/nf-core/nanostring/pull/110/files. Let me know if this would be enough for the ticket, I can move forward with finishing the subworkflow and then migrating it to nf-core
I agree, lets do this and I will also check the diffabundance solutions. There might be some silght differences there too, so maybe not ideal to try to do both in a single subworkflow.
nf-core/modules part will be addressed in this PR: https://github.com/nf-core/modules/pull/7107 nf-core/nanostring part will be addressed in this PR (once the other one gets merged): https://github.com/nf-core/nanostring/pull/114
Linting is failing with a weird error, waiting on a response from nf-core
The response I got from nf-core was that these modules weren't good candidates for migrating them to nf-core as they are not really tools but single scripts which are only hosted in the pipeline. Instead, I kept all of the work I did for standardizing the modules and subworkflow and merged that into the pipeline. The modules and new subworkflow will remain local, but they are following nf-core standards
Description of feature
Gene Score computation --> should be subworkflow within nf-core subworkflows and then be contributed to differentialabundance and/or can be used in both pipelines equivalently :)