nextflow tower add pipeline setup

[rhodesch]

For setting up pipelines on nextflow tower, can you confirm I am adding a pipeline properly? As an example, I am trying to add Cell Ranger mapping (cellranger count) pipeline from the Ingestion workflow, using google life science as a compute environment platform:

Go to Launchpad, then Add pipeline

Name: pipeline_name Pipeline to launch: https://github.com/openpipelines-bio/openpipeline Revision number: main_build <-- Is this correct? Or use main branch? Config profiles: target/nextflow/mapping/cellranger_count/nextflow.config <-- Is this needed? I've seen examples where this appears to be omitted. Main script: target/nextflow/mapping/cellranger_count/main.nf Schema name: target/nextflow/mapping/cellranger_count/nextflow_schema.json Head Job CPUs: 2 <-- Pipeline doesn't appear to build properly (dropdown not rendered upon run launch without numeric value) Head Job memory: 256 <-- Pipeline doesn't appear to build properly (dropdown not rendered upon run launch without numeric value)

Any suggestions or reproducible examples would be greatly appreciated. Thanks in advance

[DriesSchaumont]

Hi @rhodesch Thanks for reaching out!

To answer your questions:

• The main branch contains the source code only. This source code needs to be build into the nextflow and docker artifacts which can be used to actually run the pipeline. These can be found on main_build. However, the main_build branch is the development branch for this project, and changes constantly. If you would like to use a stable release, I would suggest that you use one of the tags, like 0.9.0.
• Yes, the config file is required when using OpenPipelines
• The Head Job CPUs and Head Job memory parameters are handled by nf-tower. I could be wrong, but I dont think this is pipeline-code related. In most of the usecases I have seen, the Head Job CPUs and Head Job memory was left blank because this is handled by the compute environment. However, I have no experience with using google cloud as a compute environment so YMMV. Are you following a specific guide to setup the pipeline? I found this one in the nf-tower documentation: https://help.tower.nf/21.12/compute-envs/google-cloud/#tower-configuration. I think it mentions here that you need to specify values for Head Job CPUs and/or Head Job memory in the compute environment config, which in turn populates the dropdown.

Let me know how it goes! Dries

[DriesSchaumont]

Hi @rhodesch I am going to close this issue because it has become inactive. Please reopen if you feel it is necessary