15 April 2014, Domstad, Utrecht
13:00 – 17:00
Due to the increasing number of NGS datasets and projects, there is a need for genomics core facility to build reliable, scalable, and reproducible NGS data analysis production pipelines. Furthermore, introducing a service model on top of such pipelines requires clear specification of the file system, backup policy, responsibilities, and SOPs. In this DTL focus meeting, we will discuss current practice at different groups, e.g., make, snakemake, molgenis-compute, shell scripts, etc and how to organize data analysis services. We will compare these different systems and approaches so that we can improve our local practice. Please refer to a recent initiative from Pjotr Prins for more motivation.
Speakers & subjects:
- Leon Mei (LUMC) Introduction, landscape of NGS groups, requirement of NGS production pipelines
- Wibowo Arindrarto (LUMC) Pushing the limit of GNU Make to implement NGS production pipeline
- Freerk van Dijk (UMCG) Molgenis-compute based pipelines
- Johannes Köster (TU Dortmund) Snakemake features, current status, roadmap
- Pjotr Prins (UMCU) Status, ideas, roadmap from bioinformatics open source community
- Mahdi Jaghouri (AMC) Implementing data analysis pipelines using more advanced workflow engines
The meeting will discuss advanced technical topics. The participants have to have sufficient knowledge on bioinformatics and using Linux, Cluster, Cloud/Grid to build pipelines.
Contact person for the meeting is: Leon Mei, LUMC.