From: Joshua Adelman (jadelman_at_berkeley.edu)
Date: Fri Jun 05 2009 - 10:41:44 CDT
Hi Prasun,
I think you might have some unrealistic expectations of molecular
dynamics to begin with. You should ask yourself what you expect to
see/ get out of these simulations. I would be extremely skeptical of
the thought that you would see these proteins properly bind or dock in
any reasonable amount of simulation time, when starting that far apart
and with no biasing forces. You should then ask yourself if you know
enough about the way they interact to bias their motion relative to
one another in a reasonable way.
As far as your question #7 goes, you're going to have to provide more
detail, but I suspect that this is one of the common problems that has
been addressed on the list before. I would suggest searching the list
for your specific error, and I'm sure you'll find some insightful
advice that has been already offered to others.
#8: This doesn't mean much for MD. Do you have experimental data on
the relative orientations of the two proteins when they interact, and
specific residue-residue contacts
#9: Depending on the length of your simulation before it crashes
(assuming nothing else is wrong), it is unlikely that you've simulated
long enough. Also 10 angstroms is relatively far apart and I wouldn't
expect the two proteins to have much of a sense that the other one is
there in the same simulation box
#10: You need to think carefully what you would hope to get out a
biased simulation, and what that bias would look like. These are not
trivial things to do. . .
Depending on what you are trying to accomplish, there may be much less
costly methods than MD, which may give you a better answer for your
particular case. I've only played around with these issues myself a
little bit, so there may be others on the list that have more/better
insight, but this is just my two cents for what it's worth. If you
provide more specific information about your system, people will
probably have a easier time helping you out.
Best of luck,
Josh
On Jun 5, 2009, at 12:51 AM, prasun kumar wrote:
> Dear all
>
> I am trying to study protein-protein interaction using NAMD & CHARMM
> forcefield.
>
> Steps I am following are:
>
> two proteins are placed at a distance of 10 angstrom to each other
> minimized in vacuum
> solvated in water and ionized
> Minimization was done again
> I have done the equlibration also
> while doing dynamics ,volume is getting decreased and at last
> simulation is stopped saying fatal error related to the volume and
> asking to increase the margin.
> what should I do for it????????
> I have studied that two proteins are interacting to each other(its
> proved in wetlab)
> while doing dynamics the two proteins are not seems to coming
> towards each other
> I want to know whether I should "tclforces on" or not.......
> Thanx in advance
> with regards
> Prasun kumar
>
------------------------------------------------------------------------------------------------------
Joshua L. Adelman
Biophysics Graduate Group Lab: 510.643.2159
218 Wellman Hall Fax: 510.642.7428
University of California, Berkeley http://nature.berkeley.edu/~jadelman
Berkeley, CA 94720 USA jadelman_at_berkeley.edu
------------------------------------------------------------------------------------------------------
This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:52:54 CST