This repository is a collection of institution- and cluster-specific profiles in order to provide sensible default parameters for getting your pipeline running on a particular compute setup.
The snakemake --profile <your_profile> parameter can be to load a profile
that is present in $HOME/.config/snakemake or /etc/xdg/snakemake on Linux.
In the future, you can specify a profile present in the profiles directory
of this repository, e.g. snakemake --profile dkfz and it will be automatically
loaded.
These profiles can be further augmented using --workflow-profiles for
individual workflows. These can be also searched for in profiles/default
in the working directory or next to the Snakefile.
For interacting with clusters, one of the Snakemake executor plugins can be used. This can be specified with the executor: field in config.yaml.
For further information regarding profiles, please check the relevant documentation.
Before contributing a profile, please check with the system administrator of your institution's compute system as they may want to obscure certain parameters for privacy-related reasons.
A profile should be a directory with at least two files: a config.yaml which
specifies the executor, default resources, cores, jobs, and additional
parameters, and a README.md which explains any additional instructions, e.g.
global variables to be exported in your shell profile, and names the
maintainers of the profile and their Github handles.
Please test that your profile works on your institution's compute setup. More details relating to testing will be added here shortly.
| Cluster | Platform | Location | Maintainers | Github |
|---|---|---|---|---|
| Deutsches Krebsforschungszentrum (DKFZ) | LSF | Heidelberg, DE | David Koppstein, David Laehnemann | @dkoppstein @dlaehnemann |
| BinAC 2 | HTCondor | Tübingen, DE | Felix Bartusch | @fbartusch |
| Bonn Analysis Facility (BAF) | Slurm | Bonn, DE | Slurm | @lukavom |
| Mogon | Slurm | Mainz, DE | Christian Meesters | @cmeesters |