AW: Fixing bonds in dihydrogen phosphate to save computation time

From: Norman Geist (norman.geist_at_uni-greifswald.de)
Date: Tue Jul 10 2012 - 05:36:15 CDT

Hi Axel,

> -----Ursprüngliche Nachricht-----
> Von: Axel Kohlmeyer [mailto:akohlmey_at_gmail.com]
> Gesendet: Dienstag, 10. Juli 2012 12:21
> An: Norman Geist
> Cc: Evgeny Bulat; Namd Mailing List
> Betreff: Re: namd-l: Fixing bonds in dihydrogen phosphate to save
> computation time
>
> On Tue, Jul 10, 2012 at 7:44 AM, Norman Geist
> <norman.geist_at_uni-greifswald.de> wrote:
> > Hi,
> >
> >
> >
> > I think there’s no SHAKE like feature implemented for such usage.
> It’s also
> > kind of a different simulation method, is mostly called “rigid
> body”.It is,
> > as it seems, hard to combine with all atom force field, so there are
> no
> > implementations I know in no classical md software.
>
> norman,
>
> this is an incorrect statement. codes like LAMMPS

I said I don't know one ;)

> or moldy do contain rigid body integrators.

I already heard of moldy and thought it was rigid body only. I've not found
any classical md package containing such feature. Nice to know that Lammps
has it.

>
> the problem with a rigid body integrator, is that the
> integration of the rotational degrees of freedom usually
> still requires a fairly small time step to run stable.

So it is not comfortable to use in combination as one want to use it as an
accelerating feature.

>
> axel.
>
> >
> >
> > Possibly you could use the extrabonds command to introduce some
> strong
> > forced angles/torsions/bonds to your ions to prevent them from moving
> to
> > much within, this shouldn’t effect interaction with other molecules
> and
> > could be worth a try.
> >
> >
> >
> > Regards
> >
> >
> >
> > Norman Geist.
> >
> >
> >
> > Von: owner-namd-l_at_ks.uiuc.edu [mailto:owner-namd-l_at_ks.uiuc.edu] Im
> Auftrag
> > von Evgeny Bulat
> > Gesendet: Montag, 9. Juli 2012 20:56
> > An: namd-l_at_ks.uiuc.edu
> > Betreff: namd-l: Fixing bonds in dihydrogen phosphate to save
> computation
> > time
> >
> >
> >
> > Hey everyone,
> >
> > Recently, I have explored using dihydrogen phosphate, H2PO4-, for
> ionizing
> > my biomolecular MD simulation. The reason is that H2PO4- is more
> > physiological than, say, Cl-. After generating H2PO4- toppars and
> ionizing
> > the system, I performed 10ns of equilibration and noticed that, not
> > surprisingly, the computation time is significantly slower than when
> Cl- is
> > used; in fact, it's just not feasible at the moment.
> >
> > At this point, I'm looking into somehow freezing most bonds within
> H2PO4-,
> > essentially modeling each anion as a rigid structure. I figure this
> may make
> > the computation more permissible. I have looked into how NAMD can be
> used to
> > freeze bonds between hydrogen and heavy atoms (with SHAKE), and
> wonder if a
> > similar method has been developed (or can be developed) for specific
> bonds
> > between heteroatoms.
> >
> > Thanks a bunch for any feedback.
> >
> > Best,
> > Jack Bulat,
> > Dept. of Bioengineering, UC Berkeley
>
>
>
> --
> Dr. Axel Kohlmeyer akohlmey_at_gmail.com http://goo.gl/1wk0
> International Centre for Theoretical Physics, Trieste. Italy.

This archive was generated by hypermail 2.1.6 : Tue Dec 31 2013 - 23:22:13 CST