Mr Michael Milton1,2, A/Prof Andrew Lonie1
1Melbourne Bioinformatics, University of Melbourne
2Melbourne Genomics Health Alliance
Common Workflow Language (or CWL), is a growing language for defining workflows in a cross-platform and cross-domain manner. In biology in particular, we need workflows to automate complex analyses such as DNA variant calling, RNA sequencing, and genome assembly. CWL provides a simple and well-defined format for automating these analysis by specifying their stages and connections using readable CWL documents.
CWL makes use of a number of existing standards, including support for cluster computing using SLURM or PBS, containerisation using Docker, and deployment using common packaging formats. In addition, the CWL ecosystem has grown to include workflow visualisation tools, graphical workflow editors, libraries for interacting with CWL programatically and tools that convert to and from CWL and other workflow formats.
This workshop will cover most of these facets of CWL, including:
- Wrapping tools using the CWL standard
- Writing CWL workflows
- Running CWL on different computing environments
- Using Docker containers to ensure reproducibility
- Sharing CWL tools and workflows
Use-cases for CWL in bioinformatics
Michael Milton is a software developer and bioinformatician working at Melbourne Bioinformatics. He has studied both Biology and Computer Science at university and is currently studying a Master of Data Science. Michael works for Melbourne Genomics to develop variant calling pipelines and integrate these systems with the Victorian health system, and has used a number of different pipeline frameworks, including CWL, WDL, Galaxy, DNAnexus, Luigi and Bpipe.