test_rcorrector.sh

#!/bin/bash

#SBATCH --qos bbdefault
#SBATCH --account plackarg-spl-bioinformatic-analysis
#SBATCH --ntasks 30 # request 12 cores for the job.
#SBATCH --nodes 1 # restrict the job to a single node. Necessary if requesting more than --ntasks=1
#SBATCH --time 1:00:00 # this requests 1 hour
#SBATCH --mail-type ALL 


module purge;
module load bluebear
module load bear-apps/2021b
module load Python/3.9.6-GCCcore-11.2.0
module load Rcorrector/1.0.5-GCC-11.2.0

# Define work folder
workdir="../data/test_dir"

# Removing erroneous k-mers from Illumina paired-end reads
# Run with the highest possible number of cores
SECONDS=0
echo "Running Rcorrector on raw reads" | tee -a "$workdir/pipeline_log.txt"
mkdir -p "$workdir/5_corrected"
echo "Corrected reads will be in 5_corrected folder" | tee -a "$workdir/pipeline_log.txt"

# Raw reads are in folder '4_unmapped'

fqdir="$workdir/4_unmapped"
for r1 in "$fqdir"/*1_unmapped.fastq; do
    r2="${r1%1_unmapped.fastq}2_unmapped.fastq"


    if [[ -f $r2 ]] ; then
        run_rcorrector.pl -1 "$r1" -2 "$r2" -t 24 -od "$workdir/5_corrected"
    else
        echo "$r2 not found" >&2
    fi
done
echo "Time needed to finish Rcorrector step on raw reads: $SECONDS seconds" >> "$workdir/pipeline_log.txt"