in reply to Re^5: Compare hash with arrays and print
in thread Compare hash with arrays and print

Thanks again almut!

I really appreciate your inputs.

I am trying to manipulate $_ to obtain the desired results. But when I split it with a newline the data-lines also split up. In this case, if a record holds more than one sequence data, the remaining records are not getting printed! 

while (<FASTA>)
    {
    s/^>//mg;
#    print "$_";
    my ($name) = /^([^\n]+)/;
    
    #my ($name) = /^([\w.]+)/;
      
    #print "$name\n";
        if($hash{$name} == 10){
        select FILE1;
        }
        if($hash{$name} == 20){
        select FILE2;
        }
        if($hash{$name} == 30){
        select FILE3;
        }
    #print ">$_";
    my $output = $_;
    my (@title) = split(/\n/,$output);
    print ">".$title[0]."\n".$title[1]."\n";
}

[download]

I am splitting it with a new line character, in order to print the complete header line in my output files (and I am going wrong here).

Please suggest?

Thanks!

Replies are listed 'Best First'.
Re^7: Compare hash with arrays and print
by almut (Canon) on Jul 13, 2010 at 18:23 UTC
    when I split it with a newline the data-lines also split up.

    Instead of splitting on all newlines (as split would do by default) 

        my (@title) = split(/\n/,$output);

    [download]

    you could tell split to split into two parts only: 

        my ($title, $data) = split /\n/, $output, 2;

    [download]

    In this case, $data would hold all the remaining lines of sequence data.

    That said, I'm not really sure why you want to split the record in the first place, if you then proceed to print out the parts with the same newline added in between them :)  I.e. 

        ...
    print ">".$title."\n".$data."\n";

    [download]

    would result in the same output as 

        print ">$output\n";

    [download]
[reply]
[d/l]
[select]

      Thanks once again for your input!

      When I try running my script with the following test set , it does not print anything in the output files, instead prints the data onto the screen! 

      >001.b1 gnl|ti|10009
GCTAGTGCTAGCTAGCTAGCATCGATCGAT
>002.b1 gnl|ti|10010
CAGTCAGTCGTAGTGCTAGCTGATGCTCGT
>003.b1 gnl|ti|10011
CGATCGTAGTCGTATCGATGCTGACGTAGG
>004.b1 gnl|ti|10012
CGATGCTAGTCGTAGTGCTAGTGCTATGTC
>005.b1 gnl|ti|10013
CAGTCGTAGTCGATGCTGTATCATAGCGTA
>006.b1 gnl|ti|10014
ACAGTGCTAGCTGATCGTAGCTGAGCGGAG
>007.b1 gnl|ti|10015
AGCTAGCTGATGTCGATGCTGATCGTGATG
>008.b1 gnl|ti|10016
CGATGCTGATGCTGATGCTGTAGCTATACG
>008.g1 gnl|ti|10017
GCTAGCTAGTCGTAGTCGTAGTGTCGTAGG
>009.b1 gnl|ti|10018
CGATGCTAGTCGTAGTCGTAGCTGATGCGT
>010.b1 gnl|ti|10019
CGATCGTAGTCGTAGCTGATGCTGTAGCTG
>010.g1 gnl|ti|10020
CGTAGCTGATCGTAGCGTGACTGTAGCTGG
>011.b1 gnl|ti|10021
CGTAGCTGATGCTGATCGTAGCTAGTCGAT
>011.g1 gnl|ti|10022
CAGCTGATCGTAGCTGATGCTGATGTGTGT

      [download]

      I thought maybe splitting the header and data might help. But as it turns out, I am not going the right way.

      The output I get from the script is:

      >001.b1 gnl|ti|10009
GCTAGTGCTAGCTAGCTAGCATCGATCGAT

>002.b1 gnl|ti|10010
CAGTCAGTCGTAGTGCTAGCTGATGCTCGT

>003.b1 gnl|ti|10011
CGATCGTAGTCGTATCGATGCTGACGTAGG

>004.b1 gnl|ti|10012
CGATGCTAGTCGTAGTGCTAGTGCTATGTC

>005.b1 gnl|ti|10013
CAGTCGTAGTCGATGCTGTATCATAGCGTA

>006.b1 gnl|ti|10014
ACAGTGCTAGCTGATCGTAGCTGAGCGGAG

>007.b1 gnl|ti|10015
AGCTAGCTGATGTCGATGCTGATCGTGATG

>008.b1 gnl|ti|10016
CGATGCTGATGCTGATGCTGTAGCTATACG

>008.g1 gnl|ti|10017
GCTAGCTAGTCGTAGTCGTAGTGTCGTAGG

>009.b1 gnl|ti|10018
CGATGCTAGTCGTAGTCGTAGCTGATGCGT

>010.b1 gnl|ti|10019
CGATCGTAGTCGTAGCTGATGCTGTAGCTG

>010.g1 gnl|ti|10020
CGTAGCTGATCGTAGCGTGACTGTAGCTGG

>011.b1 gnl|ti|10021
CGTAGCTGATGCTGATCGTAGCTAGTCGAT

>011.g1 gnl|ti|10022
CAGCTGATCGTAGCTGATGCTGATGTGTGT

      [download]

      My script for the same is: 

             ......
        s/^>//mg;
    my ($name) = /^([^\n]+)/;

        if($hash{$name} == 10){
        select FILE1;
        }
        if($hash{$name} == 20){
        select FILE2;
        }
        if($hash{$name} == 30){
        select FILE3;
        }
        }
    #print ">$_";
    my $output = $_;
    print ">$output\n";

      [download]

      Thanks again!

[reply]
[d/l]
[select]
        When I try running my script with the following test set , it does not print anything in the output files, instead prints the data onto the screen!

        This is most likely because with 

            my ($name) = /^([^\n]+)/;

        [download]

        you're extracting the entire header line as $name to be used as the hash key, but you have not set up your hash accordingly with keys such as "001.b1 gnl|ti|10009".  In this case, none of the select FILE1 statements would execute, so STDOUT (connected to the screen) remains the default output handle for print.

        Note that ^([^\n]+) would match everything from the beginning of the record up to the first newline, because the character class [^\n] says "match any character but a newline" (the ^ within the class negates).

        Try one of the other suggestions, or adjust the regex as needed.

[reply]
[d/l]
[select]

In Section Seekers of Perl Wisdom