How to select two files from a directory, and always every other file?-CodePudding

I am currently working on RNA-Seq data, and I have a directory containing the forward and the reverse sequences of a number of samples. I want to run tools such as SortMeRNA, but to do this I need the file names of both the forward and the reverse sequence as the data is pair-ended.

My directory looks akin to like this:

data/expression/samples/K1-01_sortmerna_trimmomatic_1.fq.gz
data/expression/samples/K1-01_sortmerna_trimmomatic_2.fq.gz
data/expression/samples/K1-02_sortmerna_trimmomatic_1.fq.gz
data/expression/samples/K1-02_sortmerna_trimmomatic_2.fq.gz
data/expression/samples/K1-03_sortmerna_trimmomatic_1.fq.gz
data/expression/samples/K1-03_sortmerna_trimmomatic_2.fq.gz
data/expression/samples/K1-04_sortmerna_trimmomatic_1.fq.gz
data/expression/samples/K1-04_sortmerna_trimmomatic_2.fq.gz
...
data/expression/samples/K1-20_sortmerna_trimmomatic_1.fq.gz
data/expression/samples/K1-20_sortmerna_trimmomatic_2.fq.gz

What I want to do is to select the files in pairs and assign them to a variable that I can then pass on to the selected software without having to make a variable for each file.

I want the code to work by giving me $FWD and $REV as the file names K1-01_sortmerna_trimmomatic_1.fq.gz and K1-01_sortmerna_trimmomatic_2.fq.gz.

The next time it iterates over the directory to give $FWD and $REV as K1-02_sortmerna_trimmomatic_1.fq.gz and K1-02_sortmerna_trimmomatic_2.fq.gz respectively.

I have made this code, which is probably not a very efficient way of dealing with this problem (and that do not work).

DATA_LOCATION=data/expression/samples/
cd $DATA_LOCATION
files=(*.fq.gz)
total=${#files[@]}
idx=0

FWD_DONE=false
REV_DONE=false

for file in "${files[@]:idx}"; do

    if [ !$FWD_DONE ]; then
        idx=$(( idx   1 ))
        FWD=$(basename $file)[$idx]
        echo $FWD
        FWD_DONE=true
        REV_DONE=false
    fi

    if [ !$REV_DONE ] && [ $FWD_DONE ]; then
        idx=$(( idx   1 ))
        REV=$(basename $file)[$idx]
        echo $REV
        REV_DONE=true
        FWD_DONE=false
    fi

    echo index $idx
done

Unfortunately, this makes the $FWD and $REV variables the same for each pass. My guess is that it has something to do with the for statement not updating with the index increment inside. Unfortunately I am very new to shell scripting, and I have yet to find any other sources that has any help.

Any help with this would be greatly appreciated! I am more than willing to trash my own code if it means that the whole process can become simpler.

CodePudding user response：

Once cded to the directory containing the data files, does this code do what you want?

for fwd in *_1.fq.gz; do
    rev=${fwd%_1.fq.gz}_2.fq.gz
    # CODE THAT USES "$fwd" AND "$rev" GOES HERE
done

See Correct Bash and shell script variable capitalization for an explanation of why I replaced FWD and REV with fwd and rev.