All Collections
Execute Pipelines
Components and Executing Versions on Datasets
Components and Executing Versions on Datasets

How to add a new component and execute on a dataset for your pipeline

Kieran McKeag avatar
Written by Kieran McKeag
Updated over a week ago

What is a component?

Components make up a Workflow and are distinct executable units that direct the actions of a Workflow. For example, in the GATK Best Practices germline variant calling workflow, the BWA or Haplotype Caller steps are individual components. To learn more about how Components and Workflows interact, see the help article here.

How to Add a New Component

From the Pipeline page, click on the Components block on the right of your screen. To add a new component, click the add button and set up your configuration before clicking submit. The new component will be added to your components list.

What is a Version?

Versions are unique states of computer software. When updates to your pipeline are made, you will want to test the performance of this new version. To do this, you will add a new version of your Component, which in turn generates a new version of the containing Workflow. You can add this new version to your Component for your Pipeline.

How to Add a New Version to your Component

Manually Adding New Component Versions

From your created Component page, click the Add button in the upper right corner of the Versions block. Here you will configure your version and populate the fields with a name, docker image URI, URL, and JSON editor. For most new versions, you will only need to update the name and docker image URI. When finished, click save to see that the version has been added to the Versions block.

Automatically Adding New Component Versions

Component versions can be added automatically via test automation triggers or docker registry watchers. For more information, see Test Automation.


How to Execute Your Version on a Dataset

To execute your Version on a dataset, go to the Datasets page, and use the checkboxes to select the datasets that you want to execute your version on. After they have been selected, click the Perform Action on Selected button found in the Datasets for Demo Variant Calling Pipeline table. Finally, click the version you want to execute from the Specific Version dropdown menu and you will kick off an execution.

Alternatively, to kick off a specific version on a single dataset, navigate to that Dataset's summary page and use the Execute Version dropdown to kick it off:

Multiple datasets can also be executed in a planned batch analysis. For more information on ways to execute your workflow versions, see the help article here.

Did this answer your question?