satsuma0.pbs

#!/bin/bash
#PBS -l nodes=1:ppn=16
#PBS -l walltime=96:00:00
#PBS -N satsuma


# Die if any variables are used that have not been declared
set -u

# Query and target species names (these will be replaced with the actual names by the build_pbs.sh script)
query="QUERY"
target="TARGET"

# Directory containing one fasta file for each entry in the query genome fasta file
datadir=split-$target

# File containing the filenames for this target that have been processed
proclist=${query}_proclist

# Create file if it does not already exist
touch $proclist

# Create output folder for the query against this target
# This folder will contain one subfolder for each entry sequence in the query genome
outdir=${query}.vs.${target}

# Make output directory if it does not already exist
mkdir -p $outdir

# Loop through every sequence in the query genome
ls $datadir/*fasta |
while read file
do
    # Get a unique id for the current file
    base=`basename $file`
    # Only process the sequence if it has not already been done
    if [[ `grep -c $file $proclist` -eq 0 ]]
    then
        rm -rf $outdir/$base
        mkdir $outdir/$base
        # Run Satsuma with default settings
        $PWD/satsuma/SatsumaSynteny \
            -t $file \
            -q data/$query.fna \
            -n 16 \
            -o $outdir/$base
        # Did the program complete successfully
        if [[ $? -eq 0 ]]
        then
            # if successful, add this filename to the completed jobs list
            echo $file >> $proclist
        fi
    else
        # if the job has already been done, tell the user we are skipping the file
        echo "skipping $file"
    fi
done

wait