Wow! This is my first post and I have to say that I really am impressed with the breadth of knowledge you guys have. The code that I have already is very similar to Godfathers (although written in python). I have used a bit of perl in the past, and I noticed you fellas were pretty good at algorithms so I decided to post here.

Thanks to the contribution by all of you, I was surprised to see Limal's link to the WebLogo as I plan on doing something similar to it for this project.

Many thanks to your idea, converting to binary is not something I would have thought of - I guess you can always learn something new.

The actual sequences being used will be protein sequences (so 20 possibilities each character instead of 4).

The purpose of this is to try to find position specific patterns in the sequence to try to estimate the binding properties of the sequences (I have additional binding score data).

In your example, I would like to find a way to display only maximal patterns. For example, in your results between sequence 1 and 2 there are 7 sub patterns (as these can all be said to be sub patterns of (.C.G....C) on line 4). When sub patterns occur on same instances, I would like to only return the maximal pattern. Thanks