the samples that you quoted (such as X-C-N-X) are generic dihedrals (X is a
wild card), so they are quite different from a specific term where all atom
types are defined. They tend to have dividers based on how many 4-atom sets
are expected to match that wild card, and also they tend to have simplified
periodicities due to symmetry and cancellation between the multiple terms
(for example, compare the O-C-N-H and X-C-N-X, where the wildcard version
would also match O-C-N-H but the specific version has an additional 1-fold
periodicity term that can't be applied to the generic. The 1-fold term
adjusts the cis/trans balance, while the 2-fold just creates a rotational
barrier but doesn't favor cis or trans).
In general you should not take a generic and substitute it for a specific
term (no wild cards) without careful thought, and adjustments.
for this particular case of o-c-N-CX, you could use the same values that
you entered for o-c-N-H, but using only the 2-fold term. Using 1,2.5, 180,2
would work out the same as what you had (4,10,180,2) since 10/4 =2.5/1, so
it wasn't wrong, but in general that kind of substitution doesn't always
work unless you carefully check.
On Thu, Feb 24, 2022 at 10:47 AM Nathan Black <nathanblack262.gmail.com>
wrote:
> Mr. Simmerling,
>
> Thank you for your suggestion.
>
> The cross-term "region" (i.e., the atoms which needed cross-term parameters
> between GAFF2 and FF14SB) participated in an amide bond with an amino acid,
> so it made sense to me that parameters from the FF14SB force field
> (parm10.dat) would be more appropriate than those from the GAFF2 force
> field (gaff2.dat). As such, I directly matched the missing parameters to
> their equivalents (or as close as could be found) in the parm10.dat file.
>
> My conversions from missing cross-term parameters to parm10.dat parameters
> are given below:
> c-N -> C-N
> o-c-N -> N-C-O
> c-N-H -> C-N-H
> c-N-CX -> C-N-CX
> c3-c-N -> CT-C-N
> o-c-N-H -> H-N-C-O
> o-c-n-CX -> X-C-N-X
> c3-c-N-H -> X-C-N-X
> c3-c-N-CX -> X-C-N-X
>
>
> The o-c-N-H dihedral in question (H-N-C-O from parm10.dat) has two
> consecutive entries, as given below.:
>
> H -N -C -O 1 2.50 180.0 -2.
> JCC,7,(1986),230
> H -N -C -O 1 2.00 0.0 1.
> J.C.cistrans-NMA DE
>
> I initially only took the first line, but now knowing the meaning of the
> negative periodicity, I took both lines into the .frcmod to define the
> o-c-N-H dihedral. This stopped the error when saving the surfactant to
> paramter topology and coordinate files.
>
> As for your question, I took the parameters directly from the X-C-N-X entry
> in parm10.dat, as given below:
>
> X -C -N -X 4 10.00 180.0 2. AA,NMA
>
> Should this "divider" term be modified beyond what is given in parm10.dat?
>
> Much appreciated,
> NDB
>
>
> On Thu, Feb 24, 2022 at 9:00 AM Carlos Simmerling <
> carlos.simmerling.gmail.com> wrote:
>
> > the problem is that you've listed a negative periodicity for o-c-N-H,
> which
> > tells it to read the next line into the same dihedral. However, your next
> > line is the o-c-N-CX. So you need to either add more terms to o-c-N-H, or
> > make the PN positive.
> > as an aside, why do you use a divider of 4 for o-c-N-CX? It's unlikely
> that
> > there are 4 sets of this term around that c-N bond.
> >
> > DIHE
> > o-c-N-H 1 2.500 180.0 -2.000
> > o-c-N-CX 4 10.000 180.0 2.000
> >
> >
> > take a look at this for format details:
> > https://ambermd.org/FileFormats.php#parm.dat
> > --
> > Carlos Simmerling, Ph.D. (he, him, his)
> > Marsha Laufer Endowed Professor of Physical and Quantitative Biology
> > Professor, Department of Chemistry
> > Associate Director, Laufer Center for Physical and Quantitative Biology
> > Room 119 Laufer Center, Stony Brook University, Stony Brook, NY
> 11794-5115
> >
> > http://www.simmerlinglab.org
> >
> > On Thu, Feb 24, 2022 at 9:50 AM Nathan Black <nathanblack262.gmail.com>
> > wrote:
> >
> > > Hello AMBER Users,
> > >
> > > I recently posted here regarding the mixing of force fields (in my
> > > specific case, GAFF2 and ff14SB) on a single residue. Thanks to the
> > > assistance of of Dr. Anselm Horn and Mr. Carlos Simmerling, I was able
> to
> > > develop a workflow that accomplishes this. However, I have run into a
> > > separate issue during its implementation that I believe warrants its
> own
> > > thread.
> > >
> > > I have written a .frcmod file that contains manually-input parameters
> at
> > > the surfactant head-tail boundary. However, when trying to save
> parameter
> > > topology and coordinate files for these surfactants in LEaP, I am given
> > the
> > > error:
> > >
> > > Error: **no torsion terms for o-c-N-CX.
> > >
> > > However, this parameter is accounted for in the .frcmod file I
> created. I
> > > suspect I have formatted the file incorrectly, but then my other manual
> > > torsion parameters are read without issue.
> > >
> > > I have attached a cleaned (single-attempt) leap.log file, as well as
> the
> > > .frcmod file I wrote. Any assistance is much appreciated!
> > >
> > > -NDB
> > > _______________________________________________
> > > AMBER mailing list
> > > AMBER.ambermd.org
> > > http://lists.ambermd.org/mailman/listinfo/amber
> > >
> > _______________________________________________
> > AMBER mailing list
> > AMBER.ambermd.org
> > http://lists.ambermd.org/mailman/listinfo/amber
> >
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Feb 24 2022 - 14:00:03 PST
