pdip

Pathogen Discovery Pipeline

clone workflow into working directory

git clone path/to/workdir
cd path/to/workdir

edit config as needed

vi config.yaml

change at least out and log folder

install dependencies into an isolated conda environment

install miniconda from https://conda.io/miniconda.html

conda env create -n pdip --file pdip.yaml 

get databases

blast

source activate pdip
cd database
mkdir blast_`date +%Y-%m-%d`
cd blast_`date +%Y-%m-%d`
perl [conda bin folder,pdip/bin]/update_blastdb.pl --decompress nt nr

activate environment

source activate pdip

execute workflow

snakemake -n -s Snakefile.py --configfile=config.yaml

example run for PBS cluster

snakemake  -p --jobs 100 \
--configfile=config.yaml \
-s Snakefile.py \
--cluster-config cluster.json \
--cluster "qsub -j eo -e  cluster.log -N {params.name} -V -l nodes=1:ppn={threads},mem={cluster.mem},walltime={cluster.walltime}" \
--keep-going --latency-wait 100 --rerun-incomplete