foreach (@file)
             {   
                 chomp; my @r = split (/\t/, $_);
                 push(@{$genes{$r[0]}{$r[3]}{$r[13]}},$r[13]); ##  Pushing a hash that includes {Gene name}{SampleIdentifier}{AA site} 
}
foreach (@file2){    # Initialize arrays w/ 0's for each gene with size of length for each gene
                chomp;
                my @r=split(/\t/,$_);
                my @zeros = (0) x $r[1]; @{$s_l{$r[0]}} = @zeros;
            }
foreach my $key1 ( sort keys %genes ) { ### this part stores the mutations with and with out Amino acid (string info) the second hash site is only numbers otherwise ignore this section 
                    foreach my $key2 ( sort keys %{$genes{$key1}} ) {
                        foreach my $key3 ( sort keys %{$genes{$key1}{$key2}} ) {
                               push(@{$AA{$key1}{$key3}},$key3);
                                my $key4=$key3;
                                $key4=~s/\D\.\D([0-9]+)\D/$1/; $key4=~s/(\*|\?|s\d+)//; 
                               push (@{$site{$key1}{$key4}},$key4);
                        }
                    }
             }

#this is the problem section
foreach my $key1 ( sort keys %AA )
              {
                   foreach my $key4 ( sort keys %{$site{$key1}} )
                        {
                         $site_length_catch{$key1}[$key4] = scalar ( @{$site{$key1}{$key4}} );
                         }    
               }