How To Extract Set Of Reads From Fastq (Or Eventually Fasta And Qual) Based On List Of Ids?

Question

[Biomonika (Noolean)]

I have a list of ids and two files (male and female). I want to find sequences and quality scores for all ids in my list.

I have:

list of ids

2 fastq files (eventually 2 fasta files and 2 qual files)

and I want to get:

extracted_sequences_based_on_ids.fastq (eventually extracted_sequences_based_on_ids.fasta and extracted_sequences_based_on_ids.qual) (can be two files that I merge afterwards)

Yes, I can write the script that will be for each id looking up the sequence and quality score and writing the result to a new file, but I would not like to reinvent the wheel in case this is something easy to perform with already existing tools:) (and might be handy for other biostars users as well)

Thanks a lot.

[Leszek]

Duplicated post: http://www.biostars.org/post/show/10353/how-to-efficiently-parse-a-huge-fastq-file/

Answer 1

[Ole Kristian Tørresen]

I think Heng Li's seqtk will do what you need: https://github.com/lh3/seqtk

Extract sequences with names in file name.lst, one sequence name per line:

seqtk subseq in.fq name.lst > out.fq

[ymwur1]

Hi, I try to extract sequences from a fastq file using the "subseq" in seqtk. But the extract file contains only the 1st sequence but no others. I am wondering whether my name.lst file does not fit with what seqtk needs. I have names of each sequence without other symbols each line in the name.lst. But the fastq file starts each sequence name with a @. Should I add @ in front of each sequence name? Or what other problem it can be?

Any suggestion is welcome. Thanks,

Chih-Ming

[Biomonika (Noolean)]

Thank you very much! It worked.

Answer 2

[Martin A Hansen]

This can be done with Biopieces and is in fact covered in the Howto.

[Biomonika (Noolean)]

Thanks a lot, I didnt know Biopieces (looks very interesting though), but I unfortunately wasnt able to install it because of some problem with Perl modul Inline.

[Martin A Hansen]

Try the Biopieces Google Group for help.