getwilds / ww-vc-trio

Variant calling WDL workflow using a trio of variant callers and analyzing consensus among them

ww-vc-trio

This WILDS WDL workflow takes samples through a standard alignment protocol and identifies variants using a trio of well-established variant callers. Finally, the workflow narrows it down to variants of confidence by analyzing consensus amongst the three variant callers.

Usage

For Fred Hutch users that are new to WDL, we recommend using PROOF to submit this workflow directly to the on-premise HPC cluster, as it simplifies interaction with Cromwell and provides a user-friendly front-end for job submission and tracking. For users outside of Fred Hutch or more advanced users who would like to run the workflow locally, command line execution is relatively straightforward:

java -jar cromwell-86.jar run ww-vc-trio.wdl \
    --inputs ww-vc-trio-inputs.json \
    --options ww-vc-trio-options.json

Although Cromwell is demonstrated here, this pipeline is not specific to Cromwell and can be run using whichever WDL execution method you prefer (miniwdl, Terra, HealthOmics, etc.).

Support

For questions, bugs, and/or feature requests, reach out to the Fred Hutch Data Science Lab (DaSL) at [email protected], or open an issue on our issue tracker.

Contributing

If you would like to contribute to this WILDS WDL workflow, see our contribution guidelines as well out our WILDS Contributor Guide for more details.

License

Distributed under the MIT License. See LICENSE for details.

ww-vc-trio's Issues

GHCR images don't allow call caching yet

Unfortunately, Docker images stored in GitHub Container Register (GHCR) aren't fully supported by Cromwell yet. The script will execute successfully, but Cromwell won't be able to identify any of the cached results during future runs. We've brought this up with the Cromwell developers, so hopefully this should get resolved relatively soon, but for now, we'll need to switch to the WILDS DockerHub registry.