Re: anisotropic pressure control

From: Maxim Igaev (maxim.igaev_at_Biologie.Uni-Osnabrueck.DE)
Date: Wed Nov 26 2014 - 05:10:09 CST

Now everything is clear! Vielen Dank, Norman!

>
> > -----Ursprüngliche Nachricht-----
> > Von: owner-namd-l_at_ks.uiuc.edu [mailto:owner-namd-l_at_ks.uiuc.edu] Im
> > Auftrag von Maxim Igaev
> > Gesendet: Mittwoch, 26. November 2014 11:05
> > An: Norman Geist; namd-l_at_ks.uiuc.edu
> > Betreff: Re: namd-l: anisotropic pressure control
> >
> > yes, this is what I want, but as far as I know setting
> > UseConstantArea to yes will allow only fluctuations along the
> > z-axis. This means that I cannot tell NAMD: "ok, set Pzz to 10 bar"
> > but rather just say "don't control Pzz in the z-direction, let it
> > fluctuate as it wants to". Or am I wrong?
> >
>
> No. Fluctuation == changing box size in Z == NPT == Pressure
> If fluctuation only happen in Z this means the pressure target is
> also only applied to Z.
>
> > In the papers that I read on anisotropic pressure control, it's
> > explicitly said that the Pzz(target), the Langevin piston target for
> > Pzz, is linearly increased/decreased. So I've been wondering, where
> > is this Pzz(target) in the NAMD config...
>
> As already meantioned, when using useconstantarea:
>
> Langevinpistontarget == Pzz
>
> The parameter langevinpiston target is also scriptable so what you
> would want is something like:
>
> Useflexiblecell yes
> Useconstantarea yes
> set Pstart 1
> set Pend = 100
> set Pstep = 1
> for {set P 0} {$P < $Pend} {incr P $Pstep} {
> langevinpistontarget $P
> run 1000000
> }
>
> >
> > Regards,
> > Maxim
> >
> > > Ah, ok. So you want to apply the pressure in Z only?
> > > That's possible by:
> > >
> > > Useflexiblecell yes
> > > Useconstantarea yes
> > >
> > > This will apply pressure only in Z direction and leave XY
> > > untouched. Is that what you want?
> > >
> > > Norman Geist.
> > >
> > >
> > > > -----Ursprüngliche Nachricht-----
> > > > Von: Maxim Igaev [mailto:maxim.igaev_at_Biologie.Uni-Osnabrueck.DE]
> > > > Gesendet: Mittwoch, 26. November 2014 10:46
> > > > An: Norman Geist
> > > > Betreff: Re: namd-l: anisotropic pressure control
> > > >
> > > > Hi Norman,
> > > >
> > > > thanks for your answer. I have a system (a microtubule) that is
> > > > effectively "infinite" in the z-dimention (via periodical
> > > > boundary conditions). The system has been equilibrated in the
> > > > NpT ensemble for several nanoseconds. Pxx, Pyy and Pzz of the
> > > > pressure tensor are nearly equal, which means the system is not
> > > > under pressure. What I'd like to do
> > > > now is to compress/stretch it in the z-dimension by manually
> > > > increasing/decreasing the Pzz component of the pressure tensor.
> > > > I know that this is possible since there have been publications
> > > > about it, in which NAMD was used, but I don't find how it's
> > > > technically done.
> > > >
> > > > langevinPistonTarget option specifies the total target
> > > > pressure, but not Pxx or Pxy, etc.
> > > >
> > > > So my questions is how can I independently control the
> > > > components
> > of
> > > > the
> > > > pressure tensor in the config file? In other words, is there
> > > > something like Langevin piston target for Pzz?
> > > >
> > > > Regards,
> > > > Maxim
> > > >
> > > > Am Wed, 26 Nov 2014 10:22:31 +0100
> > > > schrieb Norman Geist <norman.geist_at_uni-greifswald.de>:
> > > >
> > > > > If you wan't to perform NPT in all dimensions independently
> > > > > use:
> > > > >
> > > > > Useflexiblecell yes
> > > > > Useconstantarea no
> > > > > Useconstantratio no
> > > > >
> > > > > Otherwise please explain what you want to do, most people
> > > > > don't have time to follow any links and read whole papers.
> > > > >
> > > > > Norman Geist.
> > > > >
> > > > >
> > > > > > -----Ursprüngliche Nachricht-----
> > > > > > Von: owner-namd-l_at_ks.uiuc.edu
> > > > > > [mailto:owner-namd-l_at_ks.uiuc.edu] Im Auftrag von Maxim Igaev
> > > > > > Gesendet: Mittwoch, 26. November 2014 09:49
> > > > > > An: namd-l_at_ks.uiuc.edu
> > > > > > Betreff: namd-l: anisotropic pressure control
> > > > > >
> > > > > > Dear NAMD users,
> > > > > >
> > > > > > I'd like to perform anisotropic pressure control
> > > > > > simulations to stretch/compress my system using the
> > > > > > protocol published by Luan and Aksimentiev (Phys. Rev.
> > > > > > Lett., 2008). Although I understand the theory behind it
> > > > > > (Martyna et al., 1994), I haven't found
> > any
> > > > > > information on how to do this technically in the NAMD config
> > > > > > file. Am I overlooking something? The standard Nose-Hoover
> > > > > > Langevin piston pressure control allows one to only control
> > > > > > the total pressure, but not single components of the
> > > > > > pressure tensor.
> > > > > >
> > > > > > I would appreciate any help!
> > > > > >
> > > > > > Regards,
> > > > > > Maxim
> > > > > >
> > > > > > ____________________________
> > > > > >
> > > > > > Maxim Igaev, M.Sc.
> > > > > >
> > > > > > Department of Neurobiology
> > > > > >
> > > > > > University of Osnabrueck
> > > > > >
> > > > > > http://www.neurobiologie.uni-osnabrueck.de/
> > > > >
> > >
>
>

This archive was generated by hypermail 2.1.6 : Wed Dec 31 2014 - 23:23:04 CST