[gmx-users] CHARMM36 GROMACS implementation

[Mark Abraham]

Hi,

On Tue, Nov 20, 2018 at 2:45 PM John Whittaker <
johnwhittake at zedat.fu-berlin.de> wrote:

> Hi Justin, thanks for the reply and the good idea.
>
> I checked a configuration of my system using both force fields and, for
> the information of anyone who may have a similar problem in the future,
> found using gmx energy:
>
> CHARMM36 from Tom Piggot on GROMACS website:
>
> Energy                      Average   Err.Est.       RMSD  Tot-Drift
> ---------------------------------------------------------------------------
> Bond                         8026.4         --          0          0
> LJ-14                       3825.57         --          0          0
> Coulomb-14                 -66212.1         --          0          0
> LJ (SR)                    -6600.53         --          0          0
> Coulomb (SR)                -238906         --          0          0
> Potential                   -223757         --          0          0
> Kinetic En.                  101836         --          0          0
> Total Energy                -121921         --          0          0
>
> CHARMM36 official implementation from MacKerell lab website:
>
> Energy                      Average   Err.Est.       RMSD  Tot-Drift
> ---------------------------------------------------------------------------
> Bond                         8026.4         --          0          0
> LJ-14                       3825.56         --          0          0
> Coulomb-14                 -66212.1         --          0          0
> LJ (SR)                    -6600.04         --          0          0
> Coulomb (SR)                -238906         --          0          0
> Potential                   -223757         --          0          0
> Kinetic En.                  101836         --          0          0
> Total Energy                -121921         --          0          0
>
> So, from the PE/KE/E-total values, the force fields are equivalent for my
> system of DPPC and CHARMM-TIP3P. But curiously, I noticed the LJ values
>

Indeed


> are slightly different for each (LJ-14, 3825.57/3825.56) and (LJ (SR),
> -6600.53/-6600.04). Should I concern myself at all with this difference?
> Both single-point reruns were processed and ran using 5.1.2.
>

No, that's down at the level where the order of floating point operations
dominates other considerations.

Out of curiosity, I also grompp'd and reran using 5.1.5. This time I
> received the exact same LJ-14 and LJ (SR) values as in 5.1.2 (but still
> those small differences between the force field) but the overall values of
> the potential energy for both force fields was different than in 5.1.2 (by
> 14 kJ/mol).
>
> Is there a simple explanation for why this might happen?
>

Maybe. Depends in what component the difference lies, and whether any of
the release notes points to a reason - see
http://manual.gromacs.org/documentation/#older-releases

Mark

Thanks again in advance for any help.
>
> John
>
> >
> >
> > On 11/19/18 11:32 AM, John Whittaker wrote:
> >> Hi all,
> >>
> >> I am trying to reproduce the pure DPPC bilayer data found in J. Chem.
> >> Theory Comput., 2016, 12 (1), pp 405–413 (10.1021/acs.jctc.5b00935)
> >> using
> >> the recommended protocol given in the paper.
> >>
> >> Recently, I realized I made a mistake and have been using the CHARMM36
> >> forcefield provided by Tom Piggot here
> >> (
> http://www.gromacs.org/Downloads_of_outdated_releases/User_contributions/Force_fields
> )
> >> rather than the official CHARMM36 implementation found on the MacKerell
> >> lab's site (http://mackerell.umaryland.edu/charmm_ff.shtml).
> >>
> >>  From what I understand from its description in the force field
> >> documentation and the accompanying paper, the version provided by Tom
> >> Piggot should not affect my results because the lipid parameters from
> >> CHARMM36 are unchanged (while the protein parameters are) and my system
> >> consists only of water and DPPC.
> >>
> >> Am I right to assume that my results should be (theoretically)
> >> reflective
> >> of those that I would obtain from using the using the full CHARMM36
> >> force
> >> field?
> >
> > The lipids should all be the same. A single-point energy/force
> > evaluation under both force fields would confirm this.
> >
> > -Justin
> >
> > --
> > ==================================================
> >
> > Justin A. Lemkul, Ph.D.
> > Assistant Professor
> > Office: 301 Fralin Hall
> > Lab: 303 Engel Hall
> >
> > Virginia Tech Department of Biochemistry
> > 340 West Campus Dr.
> > Blacksburg, VA 24061
> >
> > jalemkul at vt.edu | (540) 231-3129
> > http://www.thelemkullab.com
> >
> > ==================================================
> >
> > --
> > 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.