Finding the mismatches

[marybrown]

i will write the complete problem i am facing. Here is the input file i am using.

sxoght: #query  hit     score   probability     qstart  qend    qorientation    tstart  tend    matches mismatches      gapOpening      gaps
@SNPSTER4_104_308EFAAXX:1:1:1694:128
GGGATAAGAGAGGTGCATGTTGGTATTTAAGGTAGT
1 alignment(s) -- reports limited to 10 alignment(s)
 
sxoght: SNPSTER4_104_308EFAAXX:1:1:1694:128     gi|122939163|ref|NM_000165.3|   -10     1.000000        1       36      +       1595    1630    35      1   00
Score = -10, P(A|R) = 1.000000
Query:          1 GGGATAAGAGAGGTGCATGTTGGTATTTAAGGTAGT 36
                  |||||||||||||||||||||||||||||| |||||
Sbjct:       1595 GGGATAAGAGAGGTGCATGTTGGTATTTAAAGTAGT 1630
 
@SNPSTER4_104_308EFAAXX:1:1:1608:94
GCAGTTTTAAGTTATTAGTTTTTAAAATCAGTACTT
14 alignment(s) -- reports limited to 10 alignment(s)
 
sxoght: SNPSTER4_104_308EFAAXX:1:1:1608:94      gi|113412254|ref|XR_018775.1|   0       0.090884        1       36      +       1578    1613    36      0   00
Score = 0, P(A|R) = 0.090884
Query:          1 GCAGTTTTAAGTTATTAGTTTTTAAAATCAGTACTT 36
                  ||||||||||||||||||||||||||||||||||||
Sbjct:       1578 GCAGTTTTAAGTTATTAGTTTTTAAAATCAGTACTT 1613

this is a big file, though the whole file looks like this. What i am trying to do exactly is to grep the header (sxoght) for display in columns and also to display where there is a mismatch in the alignment between query and sbjct. for this input file, the expected results should look like:

>gi|122939163|ref|NM_000165.3| 1595 1630 SNPSTER4_104_308EFAA +XX:1:1:1694:128 1 36 36 -10 1 1.000000 35 mismatch : 1625.GA
 
>gi|113412254|ref|XR_018775.1| 1578 1613 SNPSTER4_104_308EFAA +XX:1:1:1608:94 1 36 36 0 1 0.090884 36 mismatch : 1581.GT 1612.TG

the code which i have written is :

#!/usr/bin/perl
open(FILE,"transalign.output") or die "can not open file";
while($var=<FILE>){
        $str1=();$str2=();
        if($var=~/^sxoght:/){
                @ar=split(/\s+/,$var);
        #print ">$ar[2]\t$ar[8]\t$ar[9]\t$ar[1]\t$ar[5]\t$ar[10]\t$ar[6]\t$ar[3]\t$ar[4]\t$ar[11]\n";
        }
        if($var=~/^Query:/){
                $str1=$var;
                $str1=~s/^Query:\s+//g;
                $str1=~s/\d+\s+//g;
                $str1=~s/\s+//g;
        }
        if($var=~/^Sbjct:/){
                $str2=$var;
                $str2=~s/^Sbjct:\s+//g;
                $str2=~s/\d+\s+//g;
                $str2=~s/\s+//g;
        }
        for($i=0;$i<=length($str1);$i++)
        {
        if(substr($str1,$i,1) ne substr($str2,$i,1)){
                print substr($str1,$i,1);
                print substr($str2,$i,1);
                print "$i\n";
        }
        }
}

I am not able to use "strict and warning" because using it doesnt allow me to access the scalar variable outside the loop. In my code, i m trying to extract the positions first, so that i will subtract it from the already stored @arr values of beginning and start. I am having problems with the for loop. I know where i am going wrong, but dont know how to correct it. PLEASEEEE HELP !!!

[numberwhun]

First, there are extremely few reasons where you cannot use warnings and strict. They help alleviate the little syntactical errors you will undoubtedly experience. If you need to access variables outside of a loop where they may get populated, then you need to declare them outside of that loop.

I always have a section near the top of my script, just after all the use statements that is specifically for all variable declarations. Try that and run your code and get rid of any syntactical errors. Then, please elaborate on what you mean by "trouble with the loop". You need to be much more specific, even siting any errors.

Regards,

Jeff

[nithinpes]

Declare $str1 and $str2 outside the while loop.

my ($str1,$str2);my @ar;
while(my $line=<FILE>){ 
 

Modify the relevant section as below:

if($var=~/^Query:/){ 
                $str1=$var; 
                $str1=~s/^Query:\s+//g; 
                $str1=~s/\d+\s+//g; 
                $str1=~s/\s+//g; 
                $str2="";
        } 
        if($var=~/^Sbjct:/){ 
                $str2=$var; 
                $str2=~s/^Sbjct:\s+//g; 
                $str2=~s/\d+\s+//g; 
                $str2=~s/\s+//g; 
        } 
        if(($str1 ne "") && ($str2 ne ""))  {
              for($i=0;$i<=length($str1);$i++) 
        { 
        if(substr($str1,$i,1) ne substr($str2,$i,1)){ 
               my $add=$i+$ar[8];
               print " $add.";
               print substr($str1,$i,1);
               print substr($str2,$i,1);
                               }           
        } 
        $str1=$str2="";    print "\n"; 
}
} 
 

- Nithin