In article <bmm4lu$g7s$00$1 at news.t-online.com>, Dr Engelbert Buxbaum wrote:
> Mulatz wrote:
>>> I'm looking for software to predict protein folding.
>>>> I have the sequence of a gene and I would like to view the protein that the
>> sequence codes for. I have searched the internet and I cannot find any
>> software (commercial or otherwise) that can predict the structure.
>> Because such a thing does not realy exist (flame-prove suit doned).
>> The prediction of secondary from primary structure should, in theory, be
> possible because the entire folding information is contained in the
> primary structure (Anfinsen theorem). However, in practice and despite
> 30 years of effort, ab initio prediction (without a homologous protein
> with known structure from X-ray crystallography or NMR) results in about
> 1/3 of the amino acids being assigned wrong structures (soluble
> proteins, membrane bound is worse).
>> If you have a homologous protein with known structure the situation is
> somewhat better. But even that is sufficient only for hypothesis
> forming, not checking.
>> In short, protein structures need to be determined in vitro, not in
> silico.
Mostly true, but the success rate for 3-state secondary structure
prediction is a bit higher now for the best methods---only about 20% of
residues will be misassigned and many of those will be on the
boundaries (ends of helices or strands) where the assignment of
secondary structure will vary even after the full 3D structure is
known.
Prediction of membrane proteins is indeed quite bad---partly due to
the lack of experimental data. It seems that each new membrane
protein that is solved has required a new experimental technique!
--
Kevin Karplus karplus at soe.ucsc.eduhttp://www.soe.ucsc.edu/~karplus
life member (LAB, Adventure Cycling, American Youth Hostels)
Undergraduate and Graduate Director, Bioinformatics
Affiliations for identification only.
