Subset Fastq file from list of read numbers

Question

[johnsonn573]

I have a txt file each line of which is a number corresponding to a specific read in a fastq file. I would like to make a subsetted fastq file from my larger fastq file with just the reads corresponding to the numbers in the txt file. Is there a simple way to do this? Thank you!

[Pierre Lindenbaum]

not clear: what is that number ? the line number starting from 0 ? from 1 ? the fastq record in the file ? starting from 0 ? from 1 ?

[cpad0112]

Please post few records from input files: fastq and text. In the absense of them, i would suggest to use seqkit grep/range function @ johnsonn573

[GenoMax]

Problem is OP here only has record numbers (odd) and not fastq headers, as far as I see.

[cpad0112]

some thing like this? @ genomax @ johnsonn573. Example is with .fasta file. Same code works for fastq file and user needs to replace input fasta with input fastq. For fastq code would be: parallel seqkit range -r {}:{} test.fq :::: test.txt

input:

$ cat test.fa
>abc
atgc
>cdg
atgc
>def
atgc

$ cat test.txt 
2
3

output:

$ parallel seqkit range -r {}:{} test.fa :::: test.txt
>cdg
atgc
>def
atgc

Answer 1

[GenoMax]

Doing this by just (record?) number is going to be tricky. If you have read headers it would be much simpler to use filterbyname.sh from BBMap suite.

filterbyname.sh in=<file> in2=<file2> out=<outfile> out2=<outfile2> names=<string,string,string> include=<t/f>

names=          A list of strings or files.  The files can have one name per line, or
                be a standard read file (fasta, fastq, or sam).

Run filterbyname.sh without any options to see in-line help.

[johnsonn573]

Thank you for letting me know about this. I changed my script so that I could filter using filterbyname.sh.

Answer 2

[Pierre Lindenbaum]

assuming the number are the 1-based index of each record in the fastq.

join -t $'\t' -1 1 -2 1 \
   <(gunzip -c input.fq.gz | paste - - - - | awk '{printf("%d\t%s\n",NR,$0);}' | sort -T . -t $'\t' -k1,1 ) \
   <(sort -T . numbers.txt) | \
sort -t $'\t' -k1,1n |\
cut -f 2- |\
tr "\t" "\n" > subset.fastq

[johnsonn573]

Yes, the numbers start from 1. Thank you!

[johnsonn573]

My fastq file is not gzipped. Is there a way to modify this script so it works on an unzipped input.fq file?

[Pierre Lindenbaum]

At this point, I'm afraid you'll have to learn it by yourself.

[johnsonn573]

That's fine. I tried gzipping the fastq file, and running as input.fq.gz, but the script still wouldn't run. So there must be another problem. I will post when I have a functioning script.

[WouterDeCoster]

but the script still wouldn't run.

You'll have to be more precise if you expect any help.