Re: Print the data following few specific lines in perl

A different approach might be to use the $/ variable ( see perlvar for details ) to read in each record and then test whether it has the 'pairs' field:

use warnings;
use strict;

{
        ## modify input record separator to read
        ## each FASTA record separately
    local $/ = '>';
        
        ## test each record, printing out if it has
        ## pair field
    while (<DATA>){
        print $_ if ($_ =~ /pair:\d+/);
    }
}
__DATA__
>CVSF43567.d1 bg|346280
CGTAGCTGATGCTGTGCTGTTGTGTCGGGGGGATATATATATATATATGGGGTCGTAGTCGTAGCGCTAG
TATGCTAGCAGCGTAGATGCTGATCGATGCTGATGCTGATCGTAGTCGTAGGCTAGTGCGATCGTAGTCG
TAGTCGATGCTGATGCGTAGCTGATGTGCTGCTGATGCTAGTCGTCGTAGCTGATGCATGCTGATCGTAG
TGCTCGATGCTAGTCGTAGTCGTAGTCGTAGCGACTGATGCGATCGTAGTCGGATGCTAGCACGTAGCTG
GCTCGATGCTGATGCTGAT
>CVSF10000.x1 bg|356789 pair:789860
ATGCGTAGCTGATGTGCTGCTGATGCTAGTCGTCGTAGCTGATGCATGCTGATCGTAGTGCTCGATGCTA
GTCGTAGTCGTAGTCGTAGCGACTGATGCGATCGTAGTCGGATGATGCTGACTGATGCTGATCTGTACGT
CGTAGCTGATGCATGCGCTAGTAGCT
>CVSF10000.y1 bg|356790 pair:789859
GCTAGTCGATGCTGATGCTGTAGCTAGCGTAGTCGTACGCGCGCGCGCGCGTTTTTTGTGACGTCGTAGT
CCGTAGCTGATGCGATGCTAGTGCTGTGTCAGCTGATGTCGTGTGTAGCTGATGCTGATCGTTCGTGTGT
CGATGCTGATGCTAGTCGTAGTGTAT
[download]

In this case I am simply testing for the presence of the 'pair' field using a regex (see perlre). Hope this helps!

Just a something something...

Comment on Re: Print the data following few specific lines in perl Select or Download Code

Replies are listed 'Best First'.
Re^2: Print the data following few specific lines in perl by ad23 (Acolyte) on Jul 01, 2010 at 17:47 UTC
Thank you for your help. The code which I am trying to run - what is going wrong in there?? I am a beginner, and it would really help if I know how to modify my code to get the desired output?? Cheers!	[reply]
Re^3: Print the data following few specific lines in perl by BioLion (Curate) on Jul 05, 2010 at 08:02 UTC
It isn't really a case of right or wrong - you code is doing exactly what you are telling it to - sorting the header lines based on the descriptor field. As you will see all over the place, TIMTOWTDI (there is more than one way to do it) is a core part of Perl programming, so while your method could be made to work, it would be much more complicated than needed (you would need to buffer the squence associated with each header and print them out together, based on testing the header) and ultimately much harder to maintain etc... Just try to work out (on paper if needs be) what steps you need to achieve, than code each one of those steps - shortcuts/better/different solutions may occur to you as you work, but the important thing is that the code is clear and well documented. Maybe this doesn't answer your question, but really it is all about practice and experience, so the fact that you have made a good attempt shows that in time you'll be laughing at these sort of problems. Keep plugging away and if you get stuck again, come to the Monastery! Just a something something...	[reply]