array_sra_ascp.qsub

#!/bin/bash
#PBS -N sra_ascp_canine_n50.log                                # name of the job
#PBS -t 1-10%2		# Spawn 10 jobs, run max 2 at a time
#PBS -d /fastscratch/foo/cgp/sra_ascp
#PBS -o /fastscratch/foo/cgp/sra_ascp/sra_ascp_canine_n50.log
#PBS -e /fastscratch/foo/cgp/sra_ascp/sra_ascp_canine_n50.log
#PBS -q batch                                  # Job queue
#PBS -l walltime=24:00:00                      # Walltime in minutes
#PBS -l mem=8gb                             # Memory requirements in Kbytes
#PBS -l nodes=1:ppn=1                 # CPU reserved
#PBS -M foo@jax.org                           # for notifications
#PBS -m a                              # send email when job ends 
#PBS -r y                                         # make the jobs re-runnable
#PBS -S /bin/bash                                 # use bash shell
#PBS -V                                           # pass user env
#PBS -j oe
#PBS                                 # Any extra arguments passed onto queue

## --- DO NOT EDIT from below here---- ##
## following will always overwrite previous output file, if any.
set +o noclobber

## sleep for n seconds before running any command
FORCEWAIT=$(shuf -i 2-10 -n 1)
echo -e "\nWaiting for ${FORCEWAIT} seconds before starting workflow\n"
sleep "${FORCEWAIT}"

echo "BEGIN at $(date)"

## github repo
CODEDIR="/home/foo/pipelines/cgp_sra_upload"
## tsv file with files to be uploaded, one per row
FILELIST="${CODEDIR}/samples/files_to_upload.tsv"
export FILELIST

if [[ ! -f "$FILELIST" ]]; then
	echo -e "\nERROR: Missing table containing files to be uploaded at $FILELIST\n"
	exit 1
fi

echo "PBS_ARRAYID is $PBS_ARRAYID"
export PBS_ARRAYID

## Load aspera module
module load aspera/3.7.7

## extract path to file to be uploaded using PBS_ARRAYID
if [[ -z "${PBS_ARRAYID}" || ! -f "$FILELIST" ]]; then
	echo -e "\nERROR: Either PBS array ID is not available as ${PBS_ARRAYID}\nJob must be run in array mode\nOR\nlist of files to be uploaded is not accessible at $FILELIST\n" >&2
	exit 1
else
	echo -e "\nUsing file from: $FILELIST\nExtracting line based on PBS_ARRAYID value: $PBS_ARRAYID\n"
	## avoid quoting PBS_ARRAYID in sed command
	FILEPATH=$(sed -n ${PBS_ARRAYID}p "${FILELIST}")

	if [[ ! -z "${FILEPATH}" ]]; then
		echo -e "\nProcessing PBS_ARRAYID: $PBS_ARRAYID and FILE: $FILEPATH\n"
	else
		echo -e "\nERROR: While processing PBS_ARRAYID: $PBS_ARRAYID, found an incorrect parsing of file: $FILELIST\nParsed line from $FILELIST is: $FILEPATH\n" >&2
		exit 1
	fi
fi

#### run sra upload ####
## add SRA remote path, including subdir under user dir
"${CODEDIR}"/scripts/sra_ascp -s "subasp@upload.ncbi.nlm.nih.gov:uploads/foo@example.com_xXXYz12X/SUB0000000/" -p "${FILEPATH}"

## END ##