note AnomalousMonk In addition, the inverted character set <c> [^TAG|TAA|TGA] </C> in the OP regex does not do what I think you think it does. The | (pipe) character has no special meaning in a character set, nor does repetition of a character have significance. The set above is equivalent to the <c> [^|TAG] </C> set. My guess about what you originally intended is something along the lines of "not followed by any of the sub-sequences TAG, TAA or TGA". This can be achieved by the negative look-ahead assertion <c> (?! TAG | TAA | TGA)</c> (assuming use of the [doc://perlre#x|/x] regex modifier). However, the original expression was <c> [^TAG|TAA|TGA]? </C> (or its equivalent <c>[^|TAG]?</C>) — note the ? quantifier — meaning "Some character other than <c> | T A G </C> must be present — or not. Whatever." The ? quantifier makes the whole thing optional either in its character set form or as a negative look-ahead. On The Other Hand, the OP m// regex used the [doc://perlre#g-and-c|/g] modifier, so the intent may have been something like "step over any three bases other than a TAG, TAA or TGA sub-sequence, if present", which could be achieved by <c> (?: (?! TAG | TAA | TGA) ...)?</c> (again, with the /x modifier), but this is reading a lot into [Ekolet]'s OP. Update: Furthermore, the <c> \w+[^ATG]? </C> sub-expression is suspect. I don't understand the intention of this one either if one assumes matching against a string consisting only of A T C G characters or, more generally, only \w characters; in either case, the <c> \w+ </C> will 'consume' anything that might be matched by the optional <c> [^ATG]? </C> which is thus rendered effectless. 960195 960195