[gmx-users] simulation using tabulated dihedral potentials.

Sun Jan 5 14:56:08 CET 2014

On Sun, Jan 5, 2014 at 5:51 AM, Chandan Choudhury <iitdckc at gmail.com> wrote:

> On Sat, Dec 28, 2013 at 5:24 AM, Justin Lemkul <jalemkul at vt.edu> wrote:
>
> >
> >
> > On 12/27/13, 12:58 AM, Chandan Choudhury wrote:
> >
> >> Dear Justin and Mark,
> >>
> >> Thanks for the reply. I have regenerated the potentials for the bonded
> and
> >> non-bonded part.
> >> I observe the similar behavior as previous.
> >>
> >> Keeping dihedral switched off, the simulation proceeds. Taking in
> account
> >> the dihedral, the simulation shows the similar behavior.
> >>
> >> Now, I switched off the two dihedrals (terminal ones). Out of 5 dihedral
> >> angles, two are switched off. The simulation proceeds smoothly for
> system
> >> consisting 10-chains (120 beads) for 200000000 steps.
> >> The problem is shown for the 80-chains (960 beads).
> >>
> >> * Step           Time         Lambda        3000000    15000.00000
> >>
> >> 0.00000    Energies (kJ/mol)     Tab. Bonds    Tab. Angles      Tab.
> >> Dih.        LJ (SR)   Coulomb (SR)    1.79026e+03    3.51875e+03
> >> 6.09547e+02    4.97912e+03    0.00000e+00      Potential    Kinetic En.
> >> Total Energy    Temperature Pressure (bar)     1.08977e+04
> >> 5.78924e+03    1.66869e+04    4.84033e+02    3.57640e+03Not all bonded
> >>
> >> interactions have been properly assigned to the domain decomposition
> >> cellsA
> >> list of missing interactions:          Tab. Bonds of    880 missing
> >>  1
> >>           Tab. Angles of    800 missing      3           Tab. Dih. of
> >>  560
> >>  missing      4Molecule type 'POLCAR'the first 10 missing interactions,
> >>
> >> except for exclusions:           Tab. Dih. atoms    2    3    4    5
> >> global   566   567   568   569          Tab. Angles atoms    3    4
> >> 5      global   567   568   569           Tab. Dih. atoms    3    4    5
> >> 6 global   567   568   569   570         Tab. Angles atoms    4    5
> >> 6      global   568   569   570            Tab. Dih. atoms    4    5
> >> 6    7 global   568   569   570   571          Tab. Bonds atoms    5
> >> 6           global   569   570         Tab. Angles atoms    5    6
> >> 7      global   569   570   571            Tab. Dih. atoms    5    6
> >> 7    8 global   569   570   571
> >> 572-------------------------------------------------------Program
> >> mdrun_463, VERSION 4.6.3Source code file:
> >>
> >> /tmp/gromacs-4.6.3/src/mdlib/domdec_top.c, line: 393 Fatal error:8 of
> the
> >> 2240 bonded interactions could not be calculated because some atoms
> >> involved moved further apart than the multi-body cut-off distance (1.5
> nm
> >> )
> >> or the two-body cut-off distance (1.5 nm), see option -rdd, for pairs
> and
> >> tabulated bonds also see option -ddcheck For more information and tips
> >> for
> >> troubleshooting, please check the GROMACSwebsite at
> >> http://www.gromacs.org/Documentation/Errors
> >> <http://www.gromacs.org/Documentation/Errors>---------
> >> ----------------------------------------------*
> >>
> >>
> >>
> >> I do not understand the origin of problem. Can any one suggest how do
> the
> >> dihedrals should be taken care?
> >>
> >>
> > Without the tabulated dihedrals, the system runs.  With them, it crashes.
> >  It seems quite evident to me that whatever potential you are attempting
> to
> > apply is the direct cause of instability.  The dihedrals are either badly
> > parametrized or incorrectly implemented, but the source of the problem at
> > this point is clear.
>
>
> Thanks Justin for the response.
> According to you dihedrals are badly parameterized or incorrectly
> implemented.
>
> A. I follow the Boltzmann inversion (BI) method to get the dihedral
> potentials. I follow the steps outlined in VOTCA to obtain the tables. Here
> are summarize what I did to obtain the dihedral tables:
> 1. Using BI method, I obtain the potential from distributions.
> 2. The potentials are then smoothed and extrapolated, as implemented in
> VOTCA.
> 3. Then generate the tables.
>
> I have uploaded the tables (along with necessary files) at
> https://www.dropbox.com/s/yumzdufuys1ifdr/with-dih.tar.
>
> B. If the tabulated dihedrals are incorrectly implemented, then can you
> guide me how can this be checked and confirmed?
> Guidance regarding this would be very helpful.
>

Expanding on my suggestion several weeks ago, set up a toy system with a
geometry you know, e.g. united-atom butane with a given dihedral angle.
With a spreadsheet or something, compute what the potential and force
should be with your tabulated interaction. Then compute a
http://www.gromacs.org/Documentation/How-tos/Single-Point_Energy using your
table with mdrun. Try other configurations. Very likely you will not get
agreement and you need to work out where the table construction is
incorrect.

Mark

> Chandan
>
>
>
> >
> >
> > -Justin
> >
> > --
> > ==================================================
> >
> >
> > Justin A. Lemkul, Ph.D.
> > Postdoctoral Fellow
> >
> > Department of Pharmaceutical Sciences
> > School of Pharmacy
> > Health Sciences Facility II, Room 601
> > University of Maryland, Baltimore
> > 20 Penn St.
> > Baltimore, MD 21201
> >
> > jalemkul at outerbanks.umaryland.edu | (410) 706-7441
> >
> > ==================================================
> >
> > --
> > Gromacs Users mailing list
> >
> > * Please search the archive at http://www.gromacs.org/
> > Support/Mailing_Lists/GMX-Users_List before posting!
> >
> > * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> >
> > * For (un)subscribe requests visit
> > https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
> > send a mail to gmx-users-request at gromacs.org.
> >
> --
> Gromacs Users mailing list
>
> * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
> posting!
>
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>
> * For (un)subscribe requests visit
> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
> send a mail to gmx-users-request at gromacs.org.
>