Peptide structure modelling

Frank Fuerst ffrank at rz.uni-potsdam.de
Fri Apr 27 06:48:13 EST 2001


"Emir Khatipov" <ekhatipo at NOSPAMmidway.uchicago.edu> wrote:

>Hi everybody,
>I might have a naive question, but anyway, does anybody know if there is a
>soft (web or Windows, free) that would allow making a stick-and-ball type
>structure of an amino acid chain and suggest the most stable conformation?

I doubt if this makes sense at all.

There are some peptides that are said to adopt an alpha-helical or
beta-hairpin structure, but usually that does not mean that 99% or
even 90% is in that conformation (in H20), but only a major fraction.
And if your peptide is not designed for one of these structures, you
will find an ensemble of many many conformations, each of which has
comparable energy. Of course there is always a most stable
conformation, but that would probably mean that it is populated by,
say, 15% of the molecules, whereas the next stable conformations would
be populated by 11%, 9%, 8.5%, 5% of the molecules, and all the rest
in all other conformations.[1]

Bye, Frank

[1] I just invented these numbers, nothing special about that. Might
be quite wrong, but it shows that it doesn't make sense to talk about
the most stable one.




More information about the Proteins mailing list