Re: DIHEDral drift [2] ?

From: Brian Bennion (brian_at_youkai.llnl.gov)
Date: Wed Jun 30 2004 - 12:30:15 CDT

hello Nicholas,

I did see this message before and looked at your plots. I was hoping that
Jim Phillips would drop a line by now.
I can't remember if you send that you checked for "bad" initial dihedral
energies.

It might be one really bad interaction that is slowly working it self out
within the terms of the force field.

Did you do a "whatif" check on the initial structure?

Brian

On Wed, 30 Jun 2004, Nicholas M Glykos wrote:

>
> Dear All,
>
> Three weeks ago I posted a question about the unexpected (to us) behaviour
> of the dihedral energy term for one of our simulations. I have not
> received any feedback as yet, so I thought that I could post again the
> message this time including the corresponding NAMD script (included as
> simple text at the end of this message) and asking even for negative
> feedback (ie. if under similar conditions you have not observed anything
> like the behaviour we are seeing).
>
> Thanks,
> Nicholas
>
> ____________________________ The problem ________________________________
>
> We are running a 6 ns NPT simulation of a ~60000 atom system (using the
> latest stable NAMD version) and we observe a very slow equilibration of
> the dihedral energy term (or is it a drift ?). To make things easier to
> visualise, I've put-up a picture showing the evolution of various terms
>
> http://utopia.duth.gr/~glykos/MD_namd_list.png (88 KBytes)
>
> For the run shown we use Nose-Hoover Langevin dynamics (320K, 1atm) with a
> 2 fs step, periodic boundary conditions, PME evaluation every four
> timesteps, Shake, and (hopefully) suitable switching functions. The
> structure of the protein has been and still is fairly stable.
>
> Any insight as to why the dihedral energy term is taking so long to
> equilibrate would be greatly appreciated,
>
>
> __________________________ The NAMD script _____________________________
>
> structure ionized.psf
> coordinates heat_out.coor
> velocities heat_out.vel
> extendedSystem heat_out.xsc
> parameters parallh22x.pro
> parameters water+ions.par
>
> restartname output/restart
> restartfreq 1000
> binaryrestart yes
>
> outputEnergies 20
> outputTiming 200
> xstFreq 200
> dcdFreq 200
> wrapWater on
> wrapNearest on
> wrapAll on
>
> timestep 2
> rigidbonds all
> nonBondedFreq 2
> fullElectFrequency 4
> stepsPerCycle 8
>
> switching on
> switchDist 10
> cutoff 12
> pairlistdist 13.5
>
> Pme on
> PmeGridsizeX 128
> PmeGridsizeY 80
> PmeGridsizeZ 64
>
> exclude scaled1-4
> 1-4scaling 1.0
>
> langevin on
> langevinDamping 1
> langevinTemp 320
> langevinHydrogen on
>
> langevinPiston on
> langevinPistonTarget 1.01325
> langevinPistonPeriod 200
> langevinPistonDecay 500
> langevinPistonTemp 320
>
> useGroupPressure yes
>
> binaryoutput off
> outputname output/equi_out
>
> firsttimestep 120000
> run 3000000
>
>
>
>
> --
>
>
> Dr Nicholas M. Glykos, Department of Molecular
> Biology and Genetics, Democritus University of Thrace,
> Dimitras 19, 68100 Alexandroupolis, GREECE, Tel ++302551084036
> Fax ++302551084037, http://origin.imbb.forth.gr/~glykos/
>

*****************************************************************
**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) 424-6605 **
*****************************************************************

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:38:45 CST