shell script for concatenating fastq files

[JQL]

Hi,

I have 30 samples, each was sequenced in 3 flow cells in all 8 lanes. So I have three directories: FC1, FC2, and FC3. Inside FC1, there are 30 sub-directories: Sample_1, Sample_2,...,Sample_30. Inside Sample_1, e.g., there are 8 files for forward and 8 files for reverse (pair-end). All are .gz files.

I wrote a shell script to concatenate all the fastq reads from the same sample so that each sample has two fastq files: Sample_1_R1.fastq and Sample_2_R2.fastq (reverse).

for i in `seq 1 30`
do
gzip -dc Sample_${i}/*_R1_*fastq.gz >>Sample_${i}/Sample_${i}_R1.fastq ## forward
done

There are bugs but I can't figure out. Wonder what did I do wrong? Or someone here can show me a btter way to do this?

thanks
John

[Richard Finney]

for i in `seq 1 30`
do
cat /dev/null > bigSample_${i}.R1.fq # create empty file
cat /dev/null > bigSample_${i}.R2.fq # create empty file

# create big R1 file for i-th sample
zcat FC1/Sample_${i}/*R1*fastq.gz >> bigSample_${i}.R1.fq
zcat FC2/Sample_${i}/*R1*fastq.gz >> bigSample_${i}.R1.fq
zcat FC3/Sample_${i}/*R1*fastq.gz >> bigSample_${i}.R1.fq

# create big R2 file for i-th sample
zcat FC1/Sample_${i}/*R2*fastq.gz >> bigSample_${i}.R2.fq
zcat FC2/Sample_${i}/*R2*fastq.gz >> bigSample_${i}.R2.fq
zcat FC3/Sample_${i}/*R2*fastq.gz >> bigSample_${i}.R2.fq
done

[JQL]

Thanks Richard.
I got mine to work. But I think your code is much cleaner. I like it.

John

for i in `seq 1 30`
do
cat /dev/null > bigSample_${i}.R1.fq # create empty file
cat /dev/null > bigSample_${i}.R2.fq # create empty file

# create big R1 file for i-th sample
zcat FC1/Sample_${i}/*R1*fastq.gz >> bigSample_${i}.R1.fq
zcat FC2/Sample_${i}/*R1*fastq.gz >> bigSample_${i}.R1.fq
zcat FC3/Sample_${i}/*R1*fastq.gz >> bigSample_${i}.R1.fq

# create big R2 file for i-th sample
zcat FC1/Sample_${i}/*R2*fastq.gz >> bigSample_${i}.R2.fq
zcat FC2/Sample_${i}/*R2*fastq.gz >> bigSample_${i}.R2.fq
zcat FC3/Sample_${i}/*R2*fastq.gz >> bigSample_${i}.R2.fq
done