Re: Re: simulation becomes unstable in residue base coarse grained system

From: Fotis Baltoumas (fbaltoumas_at_biol.uoa.gr)
Date: Tue Feb 13 2018 - 06:13:14 CST

Hello,

First of all, check your system's structure, specifically, the particles
matching the numbers that give the errors.  Maybe something there is
badly modeled and causes the error.  Fix it and try again.

Having said that, I would also recommend you consider some of the
following points:

Try splitting your equilibration phase in different, successive, smaller
runs with different timestep values, starting at a small value and
slowly increasing to 10 fs (e.g. 1, 2, 3, 5 and 10 fs for five
equilibration steps). This is essentially what they do in the CHARMM-GUI
martini maker for equilibration.  If it works and you then manage to
start a production run with a 10 or 20 fs timestep with no crashes,
you're ok.

Have you tried running the equilibration at a lower temperature (300 or
310 K) instead of 340 K? I assume that you have reasons for using it,
but try lowering its value and see what happens. Higher temperature,
essentially, means extra kinetic energy and that can lead to the "atoms
moving too fast" crash.

In case your system has proteins in it, are you using any type of
restraints (e.g. elastic networks, which are often used with Martini)? 
These tend to lead to crashes, especially if they're combined with
higher temperature values.  If so, try simulating your system without
the elastic network or consider changing its force constant values.

Note that the martini protein force field includes bonds with
significantly higher constants than the others in some protein side
chains (these were put there to emulate the constraints of the original
Martini, which are available in GROMACS but not in NAMD) and, therefore,
using an elastic network or a high temperature value may be pushing your
system to its limits.

Hope I helped,
Fotis
On 02/13/2018 01:53 PM, Shahee Islam wrote:
> Comparing the namd parameters value with the published paper's value
> ,I did not get any solution. Changing the langeving damping value
> 5(according to other paper),still I am getting the same error:atomic
> velocity is too fast.Can anyone please help me how I can solve this
> problem.
> Thanking you
>
> On Feb 12, 2018 8:57 PM, "Shahee Islam" <shaheeislam726_at_gmail.com
> <mailto:shaheeislam726_at_gmail.com>> wrote:
>
> Ok.I will try.
> Thanking you
>
> On Feb 12, 2018 8:51 PM, "Giacomo Fiorin"
> <giacomo.fiorin_at_gmail.com <mailto:giacomo.fiorin_at_gmail.com>> wrote:
>
> The time needed for equilibration depends on the system, but
> 200 ns seems a long time.  There is probably something gone
> wrong in the initial setup of the system, and you should
> inspect the frames in VMD, for example looking at the atoms
> that give the error (selection by index keyword).
>
> The force field used in the RBCG tutorial is an extremely
> popular CG force field, which is easy to use in a variety of
> other codes besides NAMD. /Hint: the name of the force field
> used is in the tutorial title!/  Search the literature for
> papers that use that model.
>
> Giacomo
>
>
>
> On Mon, Feb 12, 2018 at 10:05 AM, SHAHEE ISLAM
> <islamshahee_at_gmail.com <mailto:islamshahee_at_gmail.com>> wrote:
>
> Thank you so much for your reply.Can you please tell me
> after how much time the simulation will become
> stable.Because after 200 ns still the atomic velocity is
> too fast and system becomes stable, am not getting any
> reference from any paper about the equilibration time for
> residue base coarse grained simulation in NAMD.
>
> On Feb 12, 2018 8:10 PM, "Giacomo Fiorin"
> <giacomo.fiorin_at_gmail.com
> <mailto:giacomo.fiorin_at_gmail.com>> wrote:
>
> Typically, you only need short time steps for
> equilibration, when a few of the forces are still too
> high.  After relaxing the system, you should be able
> to bring the time step back up.
>
> Giacomo
>
> On Mon, Feb 12, 2018 at 12:08 AM, Shahee Islam
> <shaheeislam726_at_gmail.com
> <mailto:shaheeislam726_at_gmail.com>> wrote:
>
> When i am doing the minimization upto higher steps
> so that the simulation runs at higher time step
> the water molecule becomes distorted.How i will
> solve the problem.
>
> On Sat, Feb 10, 2018 at 7:00 PM, Shahee Islam
> <shaheeislam726_at_gmail.com
> <mailto:shaheeislam726_at_gmail.com>> wrote:
>
> hello
> I am doing a coarse grained simulation of a
> protein in NAMD,according to residue base
> coarse grained tutorial at 340k.I have
> minimized the system 5000 steps and also i
> have checked the minimized energy values which
> converges.But the problem is,when the time
> step is 10 fs the simulation becomes unstable.
> ERROR: Atom 456 velocity is 806.693 -1734.92
> -2186.46 (limit is 1200, atom 88 of 977 on
> patch 0 pe 0)
> ERROR: Atom 457 velocity is -640.392 1941.65
> 2087.31 (limit is 1200, atom 89 of 977 on
> patch 0 pe 0)
> ERROR: Atoms moving too fast; simulation has
> become unstable (2 atoms on patch 0 pe 0).
> Can anyone please tell me what higher time
> step value is reasonable for this simulation
> because when this simulation is done at 3fs
> the simulation is going smoothly.But the time
> step 3fs is too low for coarse grained simulation.
> Thanking you
>
> Shahee Islam
> University of calcutta
> Department of chemistry
>
>
>
>
>
> --
> Giacomo Fiorin
> Associate Professor of Research, Temple University,
> Philadelphia, PA
> Contractor, National Institutes of Health, Bethesda, MD
> http://goo.gl/Q3TBQU
> https://github.com/giacomofiorin
> <https://github.com/giacomofiorin>
>
>
>
>
> --
> Giacomo Fiorin
> Associate Professor of Research, Temple University,
> Philadelphia, PA
> Contractor, National Institutes of Health, Bethesda, MD
> http://goo.gl/Q3TBQU
> https://github.com/giacomofiorin
> <https://github.com/giacomofiorin>
>

-- 
*******************************************
Fotis A. Baltoumas
Phd Candidate, Bioinformatics Postgraduate Programme
Department of Cell Biology and Biophysics
Faculty of Biology, University of Athens
Panepistimiopolis, Athens 157 01, GREECE
   --------------------------------------
email : fbaltoumas_at_biol.uoa.gr
http://biophysics.biol.uoa.gr
http://bioinformatics.biol.uoa.gr
*******************************************

This archive was generated by hypermail 2.1.6 : Tue Dec 31 2019 - 23:19:41 CST