The purpose of this code is to demonstrate using snakemake (https://bitbucket.org/johanneskoester/snakemake/overview) to make pipelines and schedule each rule as a separate job with proper dependencies and sbatch. See also this issue:

https://bitbucket.org/johanneskoester/snakemake/issue/28/clustering-jobs-with-snakemake

The Snakefile aligns reads using bwa. The file is split up into a user defined number of reads. This way the alignment of each chunk of reads can be scheduled separately.

Requirements

• Snakemake (https://bitbucket.org/johanneskoester/snakemake/overview)
• bwa (http://bio-bwa.sourceforge.net/bwa.shtml)
• samtools (http://samtools.sourceforge.net/)
• genomeCoverageBed (http://code.google.com/p/bedtools/)
• Picard (http://sourceforge.net/apps/mediawiki/picard/index.php?title=Main_Page)

Install

• Change the path to the location of sbatch_job in Snakefile-sbatch.py
• Change the location of Picard tools in Snakefile.
• Set the proper parameters in the schedule() method for your sbatch account and partition names

Test

# make sure you are in the root dir of this repo
cd test
snakemake -j 99 --debug --immediate-submit --cluster './../Snakefile-sbatch.py {dependencies}' all