[gmx-users] Confusion on improper dihedrals and Urey-Bradley (lost when converting from Amber with acpype)

[Nicolas Cheron]

Dear all,

10 days ago I had this question regarding the conversion of force fields
with acpype. After a thorough analysis (via single points between Amber and
Gromacs), I have reached the conclusion that acpype can't be use blindly
for the conversion of CHARMM22/27 force field in the Amber format (prmtop)
to Gromacs. Please note that I have found that acpype is working smoothly
when converting Amber force fields in the Amber format to Gromacs and is
very useful for that purpose. However, Urey-Bradley terms and improper
dihedrals in the CHARMM22/27 force fields are not taken into account
properly by acpype. I just wanted to report this in case other people find
the same problem.

Nicolas


2015-10-09 14:21 GMT+02:00 Alan <alanwilter at gmail.com>:

> Nicolas, if you have created your files in Amber, then the conversion to
> GROMACS should be fine. You can find some details about it in
> https://code.google.com/p/acpype/wiki/TestingAcpypeAmb2gmx
>
> Alan
>
> On 9 October 2015 at 12:39, Justin Lemkul <jalemkul at vt.edu> wrote:
>
> >
> >
> > On 10/8/15 11:48 AM, Nicolas Cheron wrote:
> >
> >> Dear all,
> >>
> >> I am joining a project where people are using the CHARMM22 force field.
> >> They converted a .prmtop file from Amber to Gromacs format with acpype.
> In
> >> the output topology, there are no Urey-Bradley potentials or improper
> >> dihedrals (for example, all the [ angle ] functions are of type 1 and
> none
> >> are 5 and all the [ dihedrals ] function are of type 9 and none are 2).
> >>
> >> I am confused, and I am wondering if something went missing when using
> >> acpype? Do you know how acpype handle this part of the amber topology?
> Or
> >> did acpype converted the Urey-Bradley and improper dihedrals from the
> >> .prmtop to other kind of functions? For an angle A-B-C with
> Urey-Bradley,
> >> I
> >> thought it would be possible to have a force as if there was a bond
> >> between
> >> A and C but I didn't find anything like that in the topology.
> >>
> >> The manual says about Urey-Bradley potential "Although this can be
> easily
> >> written as a simple sum of two terms, it is convenient to have it as a
> >> single entry in the topology file and in the output as a separate energy
> >> term" but I am not sure to understand if that means that Gromacs
> >> developers
> >> have prefered to keep it as a single entry or not.
> >>
> >>
> > Yes, it is a single entry.
> >
> > Not all U-B angles in CHARMM have an explicit 1-3 term, but if the
> > conversion is simply nixing those that exist, clearly the conversion is
> > being done wrong.  If impropers are present yet disappearing, this is
> more
> > evidence that the program is not designed to do what you want it to, or
> > that there is some problem with the input itself.
> >
> > The simplest test is to do a single-point energy in AMBER and GROMACS and
> > compare the outcome to determine if the newly generated topology is
> > sensible. If there's a mismatch, find a better method of conversion.
> >
> > -Justin
> >
> > --
> > ==================================================
> >
> > Justin A. Lemkul, Ph.D.
> > Ruth L. Kirschstein NRSA Postdoctoral Fellow
> >
> > Department of Pharmaceutical Sciences
> > School of Pharmacy
> > Health Sciences Facility II, Room 629
> > University of Maryland, Baltimore
> > 20 Penn St.
> > Baltimore, MD 21201
> >
> > jalemkul at outerbanks.umaryland.edu | (410) 706-7441
> > http://mackerell.umaryland.edu/~jalemkul
> >
> > ==================================================
> >
> > --
> > 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.
> >
>
>
>
> --
> Alan Wilter SOUSA da SILVA, DSc
> Bioinformatician, UniProt
> European Bioinformatics Institute (EMBL-EBI)
> European Molecular Biology Laboratory
> Wellcome Trust Genome Campus
> Hinxton
> Cambridge CB10 1SD
> United Kingdom
> Tel: +44 (0)1223 494588
> --
> 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.
>