Hi,

Hopefully this does the trick for you, using negative lookahead to make sure each character doesn't repeat before the comma (or end of string):

my @tst = qw(   A,G AG,CT TC,CA GAT,CGA CGAT,TG ,G
                ACGT X,A AA,G AC,GGC ATGA,TGG ATCXG,AAC
);

for (@tst) {
        my $side = qr/(?:([ACGT])(?![^,]*\g{-1}))+/;
        print $_ . (/^$side,$side$/ ? ' good' : ' bad') . $/;
}
[download]

A,G good
AG,CT good
TC,CA good
GAT,CGA good
CGAT,TG good
,G bad
ACGT bad
X,A bad
AA,G bad
AC,GGC bad
ATGA,TGG bad
ATCXG,AAC bad
[download]

Oops, more test coverage showed an issue. Needed to use relative group instead of "\1" as originally posted which had a problem with "AC,AC". Fixed.