Skip to content

Latest commit

 

History

History
59 lines (43 loc) · 3.6 KB

CONTRIBUTING.md

File metadata and controls

59 lines (43 loc) · 3.6 KB

genie-bpc-pipeline: Contributing Guidelines

Getting started

  1. Clone the repository to your local machine so you can begin making changes.

  2. On your local machine make sure you have the latest version of the develop branch:

    git checkout develop
git pull origin develop

  3. Create a feature branch off the develop branch and work on it. The branch should be named the same as the JIRA issue you are working on in lowercase (e.g., gen-1234-{feature-here}). Make sure the branch name as informative as possible.

    git checkout develop
git checkout -b gen-1234-{feature-here}

  4. Once you have made your additions or changes, make sure you write tests and run comparison scripts to ensure changes are expected.

  5. At this point, you have only created the branch locally, you need to push this to your fork on GitHub.

    git add your file
git commit -m "your commit information"
git push --set-upstream origin gen-1234-{feature-here}

  6. Create a pull request from the feature branch to the develop branch. When changes are made to script/<module_folder_name> (only applies to scripts/references and scripts/table_updates for now), a Github action will be triggered to create a docker image for the branch, you can check it here.

Nextflow Pipeline contribution

Here is how to contribute to the nextflow workflow of the genie-bpc-pipeline

Adding a new process step

If you wish to contribute a new step, please use the following guidelines:

  1. Add a new process step as a nextflow module under genie-bpc-pipeline/modules
  2. Write the process step code and add to the appropriate module folder under script/<module_folder_name>
  3. Add the process step to the workflow section in main.nf as a step
  4. Add to any pre-existing process steps that needs this step as an input and vice versa
  5. Add any new parameters to nextflow_schema.json with help text.
  6. Add any new parameter's default values to the set parameter default values section in main.nf.
  7. Add any additional validation for all relevant parameters. See validation section in main.nf

Adding a new process module

We have automated docker builds to GHCR whenever there are changes to the scripts within a "module" as each module has its own image. Whenever a new module gets added, the github workflow .github/workflows/build-docker-images.yml should be updated.

  1. Under jobs add your module name to matrix:
  2. Once you push your changes, your docker image will build and will in the form: <registry>/<repo>:<folder_name>-<branch> (Example: ghcr.io/genie-bpc-pipeline:references-gen-1485-update-potential-phi)

Default values

Parameters should be initialized / defined with default values in the set parameter default values section in main.nf

Default processes resource requirements

Defaults for process resource requirements (CPUs / memory / time) for a process should be defined in nextflow.config.

Testing

  1. To test locally, you can’t use ghcr.io/sage-bionetworks/docker_image_name locally directly. You will need to pull the docker image to local and test with it.
  2. To test on Sequra, go to the test_bpc_pipeline, edit the pipeline by pointing it to your feature branch then update. Doing this will allow you to select parameters from the dropdown menu directly.