[gmx-users] mutated structure

[Tamas Horvath]

Hi!
I want to calculate with gromacs, how certain mutations affect the structure
of a protein.

In order to do so, 1st I "mutate" the protein in the pdb file. I use a very
simple algorithm, which places the new aa in the same orientation as the old
aa was, but it does not ensure that there is no spatial clashes (though in
most cases there is no such problem).

I run gromacs simulations on this new structure (based on the demo, that
comes with it)
1st a cg+steep energy minimasition
2nd a position restraining
3rd a 20ps periodic simulated annealing, with temperatures: 300 320 320 300,
and times: 0 2 4 6
4th simple 2ps md at 300K, and I get 10 structures from this run, which then
I compare to the wt structure

What I'm asking is, that is it a reasonable approach? How may I modify it to
make it more reasonable?
How may I detect if the protein is in it's favourable conformation?
(Mutations in the core may have extensive effects, which may need more time
to "settle")

Any suggestions would be appreciated!

Hota
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20051216/84fc2f18/attachment.html>