Dumb problem with dumper.

[oxydeepu]

Hi monks,

I am here with a new query. Hope to get a quick reply.
I am try to search for the number of occurrence of 10 hexamer word in a big file of fasta sequence.
I want to get the number of occurrence of each hexamer at each position till 5000, which is the longest sequence.
the code i wrote is given below

#!/usr/bin/perl

use Data::Dumper;

open $f,$ARGV[0];

while(<$f>)
{
        chomp;
        $id = $_; #eat up the header line
        chomp($s = <$f>);
        if($s !~ /^>/)
        {
            push @seq,$s;    
        }
}

open $fa, "top_10_hexamers.txt";

while(<$fa>)
{
        chomp;
        $hx{$_}++; #hash of 10 hexamer
}

@words = sort keys %hx;

for($i = 1;$i < 5000;$i++)
{
        for($j = 0;$j <= $#words;$j++)
        {
                $result[$i][$j] = 0; # array of 5000 columns of positi
+on and words as columns 
        }
}

foreach $x(@seq)
{
                for($j = 0; $j < (length($x) - 6);$j++)
                {
                        $wrd = substr $x,$j,6;  #getting the hexamer c
+ombinations
                        foreach $w(@words)
                        {
                                if($w eq $wrd)  # comparing with the w
+ord
                                {
                                        $result->{$j}->{$wrd}++; # try
+ing to get position, word and frequency
                                }
                        }
                }
                $wrd = "";
}
print  Dumper $result;

It worked for me when my sequences where 50 bases long. I used to get a hash named $VAR1. But Now I am working with 5000, and it prints $VAR1 = undef;
Can any one help me with this problem.
Thank you in advance,

Best regards,
Deepak

[toolic]

Tip #2 from the Basic debugging checklist: start printing array and hash values as soon as you populate them:

print Dumper(\@seq);

open $fa, "top_10_hexamers.txt";

while(<$fa>)
{
        chomp;
        $hx{$_}++; #hash of 10 hexamer
}
print Dumper(\%hx);

[oxydeepu]

Thank you guys.
Sorry for spamming.
I figured it out. It works fine.
Thank you
Deepak

[aitap]

I am try to search for the number of occurrence of 10 hexamer word in a big file of fasta sequence.

Did you try using BioPerl (http://bioperl.org)?

for($i = 1;$i < 5000;$i++)
{
        for($j = 0;$j <= $#words;$j++)
        {
                $result[$i][$j] = 0; # array of 5000 columns of positi
+on and words as columns 
        }
}

If you want to initialise your array for better handling of large data, look there: Re^3: how apply large memory with perl?. By the way, this initialisation is lost because you initialize an array and work later with a hash reference. strict would never allow it.

[tobyink]

You appear to be using the $result variable for two separate purposes; here as an array of arrays:

$result[$i][$j]

and here as a hash of hashes

$result->{$j}->{$wrd}++

Make up your mind! Short of some freaky overload magic, that's not likely to work.

Meh. The first is, of course, @result. Either way, the rest of my advice still stands...

I'd strongly recommend you use strict; near the top of your script. (In particular, it's the vars feature of strict that you want, but might as well enable the whole strict pragma.) This will force you to declare the variables you use; and thus encourage you to get them straight in your own head.