http://qs321.pair.com?node_id=11103669


in reply to remove entries with duplicate characters

In general terms, the usual way to remove duplicates (which is what you're trying to do for a certain definition of duplicates) is to read sequentially your input data records and to store the values of interest (in your example the GN "TUFA" value) that you've already seen in a hash (as a hash key). Then, if the GN value has already been seen, just don't consider the entry. Else write it to output.

For example, something along these lines (untested):

my %seen; while (my $line = <$IN>) { my $gn = $1 if $line =~ / GN=(\w+) /; next if exists $seen{$1}; $seen{$1} = 1; print $line; }
That's just the general idea. Details may vary according to the circumstances.

Otherwise, if that doesn't help, I agree with 1nickt: please show the code that you have, it will be much easier to help you.