# Analysis of variables of varying numbers

i work with amino acid sequences and i want to use a selfmade model to tell me something about it, lets call it f(seq). Now i want to know the contribution of every position in the sequence onto the model. E.q. My question is what is the importance/effect of amino acid A occuring at position I in the sequence with respect to the model?

How do i visualize something like that?

I want to use my model also on several sequences of differing lengths. Somehow this throws a monkeywrench into my plans of using a neural net...

My question is simple yet i did not find anything about it. Pointers would be appreciated. Or any comment you might have. This whole idea of mine is pretty unfinished and i dont really know yet what i want. So feel free to criticize, i will update the question accordingly.

Ah and if this is the wrong place to put this here please tell me also (:

cheers and thanks

-
What kind of model do you have, in particular deterministic or probabilistic? How would you like to measure contribution? I can imagine using an independence measure if every position and $f(seq)$ are treated as random variables. If $f(seq)$ is independent of $seq[i]$ the $i$th symbol does not yield a contribution. –  Raphael Oct 7 '10 at 9:53
You are more likely to find answers at stats.stackexchange.com for this kind of question. –  András Salamon Oct 7 '10 at 10:22
the best thing would be to keep the model abstract. but it will be a probabilistic one probably. I will look into independence measures (: However i think the sequence positions should not be regarded as independent. I would like to find some pattern or sth that has a bigger contribution than a single amino acid. –  tarrasch Oct 7 '10 at 10:36
i will repost the question at stats.stackexchange and pool the answers. thanks –  tarrasch Oct 7 '10 at 10:36
Question is too vague to be answerable. –  Warren Schudy Oct 7 '10 at 20:20
add comment

## 1 Answer

"My question is what is the importance/effect of amino acid A occuring at position I in the sequence with respect to the model?"

I suggest Monotony and Surprise by Apostolico for the modeling of importance of words appearing at certain positions, or in certain patterns. Beyond that, I'm not sure what you're looking for.

-
This link is behind a paywall I think, but is a less technical explanation of the approach. springerlink.com/content/nn13140321803528/fulltext.pdf –  Aaron Sterling Oct 7 '10 at 20:18
thank you will try that. –  tarrasch Oct 8 '10 at 7:57
add comment