This repository contains a simple set of files for playing with Snakemake on the NIH Biowulf cluster. Clone this repository into a convenient place in your data directory:
git clone https://github.com/seandavi/SimpleSnakemakeTutorial.git
This will create a directory called SimpleSnakemakeTutorial that contains:
.
|-- README.md
|-- Snakefile
|-- bam
| |-- SRR015353.bam
| |-- SRR015354.bam
| |-- SRR039129.bam
| |-- SRR039130.bam
| |-- SRR039131.bam
| |-- SRR039132.bam
| |-- SRR039133.bam
| |-- SRR039134.bam
| |-- SRR039135.bam
| `-- SRR039136.bam
|-- bampost.snk
|-- config.py
|-- jobscript.sh
`-- submit.sh
Before working with snakemake, you will need to run:
module load python/3.3.2
This loads the python/3.3.2 libraries, including the snakemake executable. Note that this will also change the python executable to be python version 3.3.2. To change back, do module unload python/3.3.2.
See the snakemake documentation for more details, but a few quick commands can get you started.
snakemake -n
snakemake --summary
snakemake -n --printshellcmds
snakemake -n --dag
module load graphviz
snakamake -n --forceall --dag | dot -Tsvg > dag.svg
snakemake
snakemake cleanup
snakemake -j 8
snakemake cleanup
qsub -l nodes=1:gpfs submit.sh