finding the substring present in string and also count the number of occurrences

Could anyone tel me what is the mistake? As the program is for finding the substrings in a given string and count there number of occurrences for those substrings. but the substring must check the occurrences for every three alphabets.

for eg: String: AGAUUUAGA (i.e. for AGA, UUU, AGA)

output: AGA-2 UUU-1

print"Enter the mRNA Sequence\n";
$count=0;
$count1=0;
$seq=<>;
chomp($seq);
$p='';
$ln=length($seq);
$j=$ln/3;
for($i=0,$k=0;$i<$ln,$k<$j;$k++) {
    $fra[$k]=substr($seq,$i,3);
    $i=$i+3;
    if({$fra[$k]} eq AGA) {
        $count++;
        print"The number of AGA is $count";
    } elseif({$fra[$k]} eq UUU) {
        $count1++;
        pr开发者_运维知识库int" The number of UUU is $count1";
    }
}

This is a Perl FAQ:

perldoc -q count

This code will count the occurrences of your 2 strings:

use warnings;
use strict;

my $seq = 'AGAUUUAGA';
my $aga_cnt = () = $seq =~ /AGA/g;
my $uuu_cnt = () = $seq =~ /UUU/g;

print "The number of AGA is $aga_cnt\n";
print "The number of UUU is $uuu_cnt\n";

__END__

The number of AGA is 2
The number of UUU is 1

If you use strict and warnings, you will get many messages pointing out errors in your code.

Here is another approach which is more scalable:

use warnings;
use strict;
use Data::Dumper;

my $seq = 'AGAUUUAGA';
my %counts;
for my $key (qw(AGA UUU)) {
    $counts{$key} = () = $seq =~ /$key/g;
}
print Dumper(\%counts);

__END__

$VAR1 = {
          'AGA' => 2,
          'UUU' => 1
        };

Have a try with this, that avoids overlaps:

#!/usr/bin/perl
use strict;
use warnings;
use 5.10.1;
use Data::Dumper;

my $str = q!AGAUUUAGAGAAGAG!;
my @list = $str =~ /(...)/g;
my ($AGA, $UUU);
foreach(@list) {
  $AGA++ if $_ eq 'AGA';
  $UUU++ if $_ eq 'UUU';
}

say "number of AGA is $AGA and number of UUU is $UUU";

output:

number of AGA is 2 and number of UUU is 1

This is an example of how quickly you can get things done in Perl. Grouping the strands together as a alternation is one way to make sure there is no overlap. Also a hash is a great way to count occurrences of they key.

$values{$_}++ foreach $seq =~ /(AGA|UUU)/g;
print "AGA-$values{AGA} UUU-$values{UUU}\n";

However, I generally want to generalize it to something like this, thinking that this might not be the only time you have to do something like this.

use strict;
use warnings;
use English qw<$LIST_SEPARATOR>;

my %values;
my @spans = qw<AGA UUU>;
my $split_regex 
    = do { local $LIST_SEPARATOR = '|';
           qr/(@spans)/
         }
    ;
$values{$_}++ foreach $seq =~ /$split_regex/g;
print join( ' ', map { "$_-$values{$_}" } @spans ), "\n";

Your not clear on how many "AGA" the string "AGAGAGA" contains.

If 2,

my $aga = () = $seq =~ /AGA/g;
my $uuu = () = $seq =~ /UUU/g;

If 3,

my $aga = () = $seq =~ /A(?=GA)/g;
my $uuu = () = $seq =~ /U(?=UU)/g;

If I understand you correctly (and certainly that is questionable; almost every answer so far is interpreting your question differently than every other answer):

my %substring;
$substring{$1}++ while $seq =~ /(...)/;
print "There are $substring{UUU} UUU's and $substring{AGA} AGA's\n";