We’re moving the GATK website, docs and forum to a new platform. Read the full story and breakdown of key changes on this blog.
This section of the forum is now closed; we are working on a new support model for WDL that we will share here shortly. For Cromwell-specific issues, see the Cromwell docs and post questions on Github.
How to develop WDL workflows using Docker images locally which can be used seamlessly in the cloud
Hi Broad/WDL folks,
I love WDL and we're really excited it's potential (good work!)
One question that I have about it is about what the workflow looks like for developing WDL programs locally (on a Mac) and then deploying those on the cloud (Google). How do you do this at Broad?
From here (https://docs.google.com/document/d/1jv4FFWC0RJUWoQagh7wA7L7K6lHNz72zGttwEfVKgnA/edit?ts=5601ad23#heading=h.16u3o9piptvm) it seems that you work within a VM when making Docker images on a Mac (although this is no longer required in the latest Docker release - https://docs.docker.com/engine/installation/mac/ - is this required if I want to call containers inside a WDL program? Said another way, is it possible to use containers made on a Mac within a WDL program from the standard Mac terminal? In turn, are there substantial changes to these programs needs before they can be deployed on a cluster?
In general, any help using WDL to write pipelines using Docker containers holding needed functions (bwa, picard, GATK) that can be developed and executed locally on a mac and ported easily to a cluster computing environment would be really helpful!