#!/usr/bin/perl

# This script reads in a nt file in fasta format, concatenates
# it all into the variable $str and then extracts out "genes" 
# by finding ATG and then the next in-frame TAA

while (<>){
    if (/^>/) { next; }
    chomp;
    $str = $str.$_;
}

while ($str =~ /(ATG([ACGT]{3})+?TAA)/g){
    print "$1\n";
}