concatenate_fast1.sh

#!/bin/bash

# author: Patricia Agudelo-Romero, PhD
# email: p.agudeloromero@gmail.com
# Concatenate FASTQ files split by lanes (L00*) into one per sample.
# It does the same for R1 and R2

if [ $# -lt 2 ]; then
    echo "Usage: $0 <input_dir> <output_dir> [jobs]"
    exit 1
fi

INPUT_DIR=$1
OUTPUT_DIR=$2
JOBS=${3:-4}   # default 4 parallel jobs

mkdir -p "$OUTPUT_DIR"

# Generate commands
> concat_commands.txt
for R1 in "$INPUT_DIR"/*L00*_R1_001.fastq.gz; do
    SAMPLE=$(basename "$R1" | sed -E 's/_L00[0-9]_R1_001.fastq.gz//')

    echo "sample: $SAMPLE"
 

    # Output filenames
    OUT_R1="${OUTPUT_DIR}/${SAMPLE}_R1_001.fastq.gz"
    OUT_R2="${OUTPUT_DIR}/${SAMPLE}_R2_001.fastq.gz"

    # Write commands for R1 and R2
    echo "cat ${INPUT_DIR}/${SAMPLE}_L00*_R1_001.fastq.gz > ${OUT_R1}" >> concat_commands.txt
    echo "cat ${INPUT_DIR}/${SAMPLE}_L00*_R2_001.fastq.gz > ${OUT_R2}" >> concat_commands.txt
done

# Run in parallel
parallel -j "$JOBS" < concat_commands.txt

#rm concat_commands.txt