Sorry I didn't see the earlier message. If you can send me the topology
then I can take a look. The topology is what ultimately matters, but this
may also indicate that there is another
problem right behind what you are seeing, that perhaps the torsions are not
present in the GAFF2 file and will need to be supplemented with a frcmod
file in the &files namelist. Another possibility is that the GAFF2
parm-file format contains some things that create issues (parm is a
remarkably ill-defined format and things can happen which make mdgx think
that certain sections are done when they are really not).
Dave
On Mon, Jan 10, 2022 at 2:52 AM Casalini Tommaso <
tommaso.casalini.chem.ethz.ch> wrote:
> Dear Amber users and developers,
>
> I am reposting my request since unfortunately I did not receive any
> feedback.
>
> I would like to use mdgx to refine the estimate of the force field terms
> related to the torsional barriers along the backbone of a polymer. I am
> using a tetramer as reference system.
>
> I created a library with monomer units, properly assigning atom types,
> charges, head and tail atoms.
>
> I am following the mdgx tutorial available on Amber website. In
> particular, I try to fit the terms related to three different torsions. In
> terms of gaff2 atom types, they are the following:
>
>
> c3 c os c3
>
> c os c3 c
>
> os c3 c os
>
>
> I computed the energies of about 500 structures and I have successfully
> created the files containing the coordinates and the energies. When I try
> to perform the fit, I receive the following message:
>
>
> mdgx >> Unique adjustable parameters enumerated.
> mdgx >> 564 conformations read.
> mdgx >> Parameter keys read.
> mdgx >> Torsion type os c3 c os marked for optimization but not
> found
> mdgx >> in any systems.
>
>
> mdgx >> Torsion type c os c3 c marked for optimization but not
> found
> mdgx >> in any systems.
>
> Essentially, the code says that two out three torsions are not present,
> but they are. I did another check, creating a mol2 file of the tetramer and
> using parmchk to print the entire parameter set (using the option -a 'Y').
> The resulting .frcmod file confirms that the torsions are indeed present.
> This is the input file for fitting (I copied gaff2.dat in the directory
> where I am performing the fitting):
>
> &files
> -parm gaff2.dat
> -d PLL.dat
> -o fit.out
> &end
>
> ¶m
> System PLL.top coords.cdf energies.dat
> ParmOutput frcmod
> eunits hartree
> accrep report.m
> verbose 1
>
> fith os c3 c os
> fith c3 c os c3
> fith c os c3 c
> hrst 0.0002
>
> &end
>
> Am I missing something?
>
>
> I thank in advance for the support.
>
> Tommaso
> _______________________________________________
> 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 Mon Jan 10 2022 - 10:00:02 PST
