Skip to content

Is it possible to use looper link on a generic pipeline? #550

@nsheff

Description

@nsheff

The docs say looper link requires pipestat to be set up.

The thing is, I don't think it should require pipestat... as long as there's an output schema, I think I should be able to use looper link.

Use case:

I have a simple pipeline interface to run cellranger-atac across samples:

sample_interface:
  command_template: >
    cellranger-arc count \
      --id={sample.sample_name} \
      --reference={project.refpath} \
      --libraries="{looper.piface_dir}/{sample.library_sheet}" \
      --localcores={compute.cores} \
      --localmem={compute.mem}

nothing complicated. the pipeline is not made by me, it's using a general third-party tool. So, obviously it doesn't use pipestat.

I feel I should now be able to:

  1. create an output schema that describes the outputs of this pipeline.
  2. attach that output schema somewhere
  3. run looper link

Maybe you can do this but I couldn't find any documentation about it:

https://pep.databio.org/looper/how-to/link-objects/

Here it says:

You must configure pipestat to use this feature.

Here it doesn't discuss it, as this tutorial is really focused on using pipestat to report results:

https://pep.databio.org/looper/user-tutorial/user-pipestat/

I suggest:

  • Add a bit about how looper link works to the pipestat user tutotiral
  • Increase docs on the looper link page
  • allow it to work with just an output schema

Metadata

Metadata

Assignees

No one assigned

    Labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions