How can I extract the start and end codon from DNA sequences in Perl?

Question

I have a code below that tries to identify the position of the start and end codons of these DNA sequences. We define the start codon as an ATG sequence and the end codon as TGA, TAA, TAG sequences .

The problem is that the code below only works for the first two sequences (DM208659 and AF038953), but not for the rest.

What happened to my approach below?

This code can be copied from here .

      #!/usr/bin/perl -w


while (<DATA>) {
    chomp;
    print "$_\n";
    my ($id,$rna_sq) = split(/\s+/,$_);

    local $_ = $rna_sq;
    while (/atg/g) {
        my $start = pos() - 2;

        if (/tga|taa|tag/g) {

            my $stop    = pos();
            my $gene    = substr( $_, $start - 1, $stop - $start + 1 ),$/;
            my $genelen = length($gene);
            my $ct      = "$id $start $stop $gene $genelen";
            print "\t$ct\n";

        }

    }

}

__DATA__
DM208659    gtgggcctcaaatgtggagcactattctgatgtccaagtggaaagtgctgcgacatttgagcgtcac
AF038953    gatcccagacctcggcttgcagtagtgttagactgaagataaagtaagtgctgtttgggctaacaggatctcctcttgcagtctgcagcccaggacgctgattccagcagcgccttaccgcgcagcccgaagattcactatggtgaaaatcgccttcaatacccctaccgccgtgcaaaaggaggaggcgcggcaagacgtggaggccctcctgagccgcacggtcagaactcagatactgaccggcaaggagctccgagttgccacccaggaaaaagagggctcctctgggagatgtatgcttactctcttaggcctttcattcatcttggcaggacttattgttggtggagcctgcatttacaagtacttcatgcccaagagcaccatttaccgtggagagatgtgcttttttgattctgaggatcctgcaaattcccttcgtggaggagagcctaacttcctgcctgtgactgaggaggctgacattcgtgaggatgacaacattgcaatcattgatgtgcctgtccccagtttctctgatagtgaccctgcagcaattattcatgactttgaaaagggaatgactgcttacctggacttgttgctggggaactgctatctgatgcccctcaatacttctattgttatgcctccaaaaaatctggtagagctctttggcaaactggcgagtggcagatatctgcctcaaacttatgtggttcgagaagacctagttgctgtggaggaaattcgtgatgttagtaaccttggcatctttatttaccaactttgcaataacagaaagtccttccgccttcgtcgcagagacctcttgctgggtttcaacaaacgtgccattgataaatgctggaagattagacacttccccaacgaatttattgttgagaccaagatctgtcaagagtaagaggcaacagatagagtgtccttggtaataagaagtcagagatttacaatatgactttaacattaaggtttatgggatactcaagatatttactcatgcatttactctattgcttatgccgtaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
BC021011    ggggagtccggggcggcgcctggaggcggagccgcccgctgggctaaatggggcagaggccgggaggggtgggggttccccgcgccgcagccatggagcagcttcgcgccgccgcccgtctgcagattgttctg
DM208660    gggatactcaaaatgggggcgctttcctttttgtctgtactgggaagtgcttcgattttggggtgtccc
AF038954    ggacccaagggggccttcgaggtgccttaggccgcttgccttgctctcagaatcgctgccgccatggctagtcagtctcaggggattcagcagctgctgcaggccgagaagcgggcagccgagaaggtgtccgaggcccgcaaaagaaagaaccggaggctgaagcaggccaaagaagaagctcaggctgaaattgaacagtaccgcctgcagagggagaaagaattcaaggccaaggaagctgcggcattgggatcccgtggcagttgcagcactgaagtggagaaggagacccaggagaagatgaccatcctccagacatacttccggcagaacagggatgaagtcttggacaacctcttggcttttgtctgtgacattcggccagaaatccatgaaaactaccgcataaatggatagaagagagaagcacctgtgctgtggagtggcattttagatgccctcacgaatatggaagcttagcacagctctagttacattcttaggagatggccattaaattatttccatatattataagagaggtccttccactttttggagagtagccaatctagctttttggtaacagacttagaaattagcaaagatgtccagctttttaccacagattcctgagggattttagatgggtaaatagagtcagactttgaccaggttttgggcaaagcacatgtatatcagtgtggacttttcctttcttagatctagtttaaaaaaaaaaaccccttaccattctttgaagaaaggaggggattaaataattttttcccctaacactttcttgaaggtcaggggctttatctatgaaaagttagtaaatagttctttgtaacctgtgtgaagcagcagccagccttaaagtagtccattcttgctaatggttagaacagtgaatactagtggaattgtttgggctgcttttagtttctcttaatcaaaattactagatgatagaattcaagaacttgttacatgtattacttggtgtatcgataatcatttaaaagtaaagactctgtcatgcaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa

Answer 1

$_ ( , local - , , - clobber $_, $rna_sq, ?

, $start $stop 0 ( ) $genelen , substr . ( $[ 1, 1, . perldoc perlvar.)

use strict;
use warnings;
while (my $line = <DATA>) {
    chomp $line;
    print "processing $line\n";
    my ($id, $rna_sq) = split(/\s+/, $line);

    while ($rna_sq =~ /atg/g) {
        # $start and $stop are 0-based indexes
        my $start = pos($rna_sq) - 3; # back up to include the start sequence

        # discard remnant if no stop sequence can be found
        last unless $rna_sq =~ /tga|taa|tag/g;

        my $stop    = pos($rna_sq);
        my $genelen = $stop - $start;

        my $gene    = substr($rna_sq, $start, $genelen);
        print "\t" . join(' ', $id, $start+1, $stop, $gene, $genelen) . "\n";
    }
}

Answer 2

, if (/tga|taa|tag/g) . /atg/g , . :

if (/tga|taa|tag/g) {
    ...
}
else {
    last;
}

Answer 3

, , . , AF038954 atgaccatcctccagacatacttccggcagaacagggatga, atgaagtcttggacaacctcttggcttttgtctgtga. ?

, , , :

while (<DATA>) {
    chomp;
    print "processing $_\n";
    my ($id, $rna_sq) = split;

    while ($rna_sq =~ /(atg.*?(?:tga|taa|tag))/g) {
      printf "\t%8s %4i %4i %s %i\n",
             $id,
             pos($rna_sq) - length($1) + 1,
             pos($rna_sq),
             $1,
             length($1);
      }
}

(atg.*?(?:tga|taa|tag)) , , ( ? .* "" ), . while , .

, , : , , , , , . :

while (<DATA>) {
    chomp;
    print "processing $_\n";
    my ($id, $rna_sq) = split;

    while ($rna_sq =~ /atg/g) {
      if ($' =~ /(.*?(?:tga|taa|tag))/) {
        my $match = "atg$1";
        printf "\t%8s %4i %4i %s %i\n",
               $id,
               pos($rna_sq) - 2,
               pos($rna_sq) - 3 + length($match),
               $match,
               length($match);
      }
    }
}

Here we use a special variable (usually not recommended) $'that contains content after the match. We will look at this to find the end of the sequence and bring out the details. Since our main global match with $rna_seqdoes not include the sequence (as indicated above), we restart the search for the beginning, in which the previous search stopped, which was immediately after the start of the search. Thus, we include overlapping sequences.