If the files are very large, you don't want to load the whole file into memory. A more efficient solution would be to only read the last (10 + epsilon) characters of the file (the epsilon "fudge factor" is just in case there is a \r or \r\n within the last 10 characters.

Here is some code I came up with to do this search efficiently:

my $tail_length = 10;

sub contains_polytail {
    my $file = shift;
    open my $fh, "<$file" or die "Couldn't open $file: $!";

    ## only read the last 10 + 2 bytes

    seek $fh, -( $tail_length + 2), 2;
    my $tail = do { local $/; <$fh> };

    ## ignore newlines and ^M's
    $tail =~ s/[\r\n]//g;

    return ($tail =~ /[AN]{$tail_length}$/);
}

print contains_polytail('test.dna') ? "yes\n" : "no\n";
[download]

Important note: this won't tell you the entire polyA tail, just whether there is one. The polyA tail might be 600 characters, it might be 15. If you need the entire tail's contents, you either have to keep reading backwards through the file, or else simply slurp the entire file into memory and use one of the other solutions in this thread. But now you only need to slurp the entire file if you know it contains a polyA tail.

blokhead