Fastq Extract Under-Represented Sequences

Question

[daniel.soronellas]

Hi everyone,

I have a fastq file sample which has almost 57% of over-representation of different sequences related to a Linker which means that something went wrong when the library was performed. But above this, I want to retrieve the almost 40% left to perform a FASTQc analysis and know if library preparation with the other linkers went properly.

So, do you know if there is any tools or even any script that could extract sequences that not correspont to the linker which is giving me this over-representation?

The linker that went wrong has the following sequence: TCGTAT.............TTG

Thanks,

Answer 1

[JC]

If the linker is expect to be always as the first bases in your read sequence, you can use a Perl script like this:

#!/usr/bin/perl

use strict;
use warnings;

my $nl = 0;
my $match = 'TCGTAT.............TTG';
my ($read, $seql);

while (<>) {
    $nl ++;
    $read .= $_;
    if ($nl == 2) { 
        $seq = $_; 
    }
    elsif ($nl == 4) { 
         print $read unless ($seq =~ m/^$match/);
         $read = '';
         $nl = 0;
    }
}

then you can run:

perl filter.pl < file.fq > filtered.fq