From: Brian Bennion (brian_at_youkai.llnl.gov)
Date: Thu Nov 06 2003 - 11:16:03 CST
Hi Himanshu
At this point I don't understand what is happening in the first case or
the second case. But it might be helpful just to solvate the first protein
with a 8A layer of water to see if you can get past the 200 steps of
dynamics which I still think are unorthodox.
Imagine a tree at 0 Kelvin, now in 1fs imagine a tree at 300 Kelvin.
The energy has to be distributed according to a maxwellian or other
distribution to make the overall temperature 300K.
Some atomsare going to get a pulse that is greater than the equivalent of 300K while other
atoms will be a bit colder. A single pulsed heating event could send some
"borderline bad" interactions over the edge and the simulation
breaks down.
For the second case if you could use a debugger to look at the core file.
ie
gdb namd2.5 XXXX.core.XXX
and execute the command
where
then send the results to the list
Brian
On Thu, 6 Nov 2003, himanshu
chandola wrote:
> hi,
> well i had my system's initial temperature at 300K and I had langevin
> dynamics in place with temperature at 300K. So i didn't heat up my
> protein.I didn't solvate my protein because the dynamics simulations which
> i intend to do will not be much changed by inclusion of water (which will
> increase the computation time). I also did notice the difference between
> namd and vmd numbering but the atoms which were moving fast - didn't
> suggest something meaningful.
>
> In the meanwhile - i have this second problem with a protein which I am
> minimizing over 10000 steps. I get a message and namd exists with the
> following message . any clues? and thank you all for being supportive
> especially Brian.
>
> LDB: LOAD: AVG 439.251 MAX 703.499 MSGS: TOTAL 48 MAXC 24 MAXP 4 None
> LDB: LOAD: AVG 439.251 MAX 527.099 MSGS: TOTAL 80 MAXC 28 MAXP 4 Alg7
> LDB: LOAD: AVG 439.251 MAX 448.024 MSGS: TOTAL 81 MAXC 28 MAXP 4 Alg7
> Signaled (core dumped).
>
> This occurs after approx 3 minutes into the minimization.
>
>
> himanshu
>
> ----------------------------------------
> Morpheus: Do you believe in fate, Neo?
> Neo: No.
> Morpheus: Why Not?
> Neo: Because I don't like the idea that I'm not in control of my life.
>
>
> On Wed, 5 Nov 2003, Brian Bennion wrote:
>
> > Hello,
> >
> > Just thinking outloud...
> > It is not common to equilibrate a protein without some sort of solvation
> > unless you mean you are minimizing the protein, but that would not
> > generate the error you see.
> >
> > You didn't answer the question about how you are heating the protein or
> > what temperature you are trying to equilibrate to. Again it is strange
> > that there is no solvent in your equilibration.
> >
> > Remember that the atom numbers in namd need to have one subtracted from
> > them to see the same atom in VMD. Are the atoms in question internal or
> > on the surface of the protein?
> >
> > Again, with out some kind of idea what temperatures you are at and what
> > dielectric screening you are using, you could be getting some rather weird
> > charge interactions..
> >
> > Brian
> >
> >
> > On Thu, 6 Nov 2003, himanshu chandola wrote:
> >
> > > hi,
> > > i am using a time step of 1fs. I am not solvating my protein - that's
> > > the particular reason i didn't use a box. Increasing the cutoff just made
> > > the 'global exclusion error' which was coming alongwith vanish but still
> > > namd crashes after some odd 199 or some timesteps with the message: atoms
> > > moving too fast. I also looked into the atoms which were cited as moving
> > > fast and there wasn't anything peculiar in them.
> > >
> > > Could it be that there is some 'wrong' generation done of my psf and pdb
> > > files - but then the structures are similar and there is no error at the
> > > first time step itself.
> > >
> > > ----------------------------------------
> > > Morpheus: Do you believe in fate, Neo?
> > > Neo: No.
> > > Morpheus: Why Not?
> > > Neo: Because I don't like the idea that I'm not in control of my life.
> > >
> > >
> > > On Wed, 5 Nov 2003, Brian Bennion wrote:
> > >
> > > > Hello,
> > > >
> > > >
> > > > What is your timestep? 2.0fs?
> > > > Is your protein solvated? What atoms are moving too fast? How fast are you
> > > > heating it up? 1 degree every 100 steps?
> > > >
> > > > Increasing your cutoff may or may not make things better. If you don't
> > > > have a box for the system you could be accelerating water atoms off into
> > > > space at light speed...
> > > >
> > > >
> > > >
> > > > On Thu, 6 Nov 2003, himanshu chandola wrote:
> > > >
> > > > > hello everyone,
> > > > > I had mailed here earlier about the error about atoms moving
> > > > > too fast. I increased my cutoff to 15 just to make sure that errors don't
> > > > > appear dude to the reason specified at
> > > > > http://www.ks.uiuc.edu/Research/namd/current/ug/node23.html
> > > > >
> > > > > Still the 'atoms moving too fast' error persists. I also don't have any
> > > > > atoms at 'uninitialized --(0,0,0)' coordinates.
> > > > >
> > > > > My protein is not in a box and at this moment i was just trying to
> > > > > equilibrate it.
> > > > >
> > > > > any suggestions about what could have been possibly gone wrong ?
> > > > >
> > > > > with kind regards,
> > > > >
> > > > > himanshu
> > > > >
> > > > > ----------------------------------------
> > > > > Morpheus: Do you believe in fate, Neo?
> > > > > Neo: No.
> > > > > Morpheus: Why Not?
> > > > > Neo: Because I don't like the idea that I'm not in control of my life.
> > > > >
> > > >
> > > > --
> > > > *****************************************************************
> > > > **Brian Bennion, Ph.D. **
> > > > **Computational and Systems Biology Division **
> > > > **Biology and Biotechnology Research Program **
> > > > **Lawrence Livermore National Laboratory **
> > > > **P.O. Box 808, L-448 bennion1_at_llnl.gov **
> > > > **7000 East Avenue phone: (925) 422-5722 **
> > > > **Livermore, CA 94550 fax: (925) 422-6605 **
> > > > *****************************************************************
> > > >
> > > >
> > > >
> > >
> >
> > --
> > *****************************************************************
> > **Brian Bennion, Ph.D. **
> > **Computational and Systems Biology Division **
> > **Biology and Biotechnology Research Program **
> > **Lawrence Livermore National Laboratory **
> > **P.O. Box 808, L-448 bennion1_at_llnl.gov **
> > **7000 East Avenue phone: (925) 422-5722 **
> > **Livermore, CA 94550 fax: (925) 422-6605 **
> > *****************************************************************
> >
> >
> >
>
-- ***************************************************************** **Brian Bennion, Ph.D. ** **Computational and Systems Biology Division ** **Biology and Biotechnology Research Program ** **Lawrence Livermore National Laboratory ** **P.O. Box 808, L-448 bennion1_at_llnl.gov ** **7000 East Avenue phone: (925) 422-5722 ** **Livermore, CA 94550 fax: (925) 422-6605 ** *****************************************************************
This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:37:06 CST