Toby Skinner wrote:
> However I have little knowledge of proteins and the last time I talked
> about this with a biologist, he said that there was no point because
> the secondary structure of a protein is context sensitive such that you
> can only get about a 70% correct structure by looking only at the
> amino-acid sequence (and not at the spatial relationship between
> helices, beta sheets etc).
Right.
> If your this far then thankyou for sticking with me, what I need is a
> set of rules for predicting the secondary structure given the primary
> structure. The idea being that an algorithm for predicting structures
> 70% correct is better than nothing, also I can then start on a new
> algorithm for applying a different set of rules for folding the
> secondary structure into the tertiary structure.
>> I understand that I am grossly simplifying this but I have spent a long
> time on this dissertation and I would really like to sort this problem
> out.
Are you familiar with other attempts on structure prediction? It
seems, not too well.
I feel you should go a bit into biology and biologists' structure
prediction literature. For a starting point, try "CASP" in PubMed.
Frank
--
Potsdam University, Germany
No courtesy copies!
Bitte keine Kopien per Mail!