Hi Dave,
I am working with Qi on this project and will attempt to explain what we
are trying to do. We are placing dummy atoms at specific points around a
curved lipid bilayer, to serve as a "wall" or physical guide to keep the
curvature in place. The curvature is to be maintained through Lennard-Jones
repulsion between the dummy atoms and the carbon atoms of the lipid
molecules.
We originally defined new atom types, AN and SH, to serve as dummy
particles. (The AN particles were to be immobile and noninteracting, while
the SH particles were to be slightly mobile and LJ repulsive.) We gave them
an atomic number of -1, which seems to have signaled to Amber (pmemd.cuda)
that we were using "extra points" (EP). After receiving your email, we
tried changing the dummy atom names to EP, but tleap recognized them as
solvent hydrogen atoms detached from the (expected) solvent molecules.
Our strategy to achieve selective Lennard-Jones repulsion between the dummy
particles and the lipid carbons was to use Parmed to define a new
Lennard-Jones type and to define a pairwise Lennard-Jones potential between
the dummy particles (SH) and the lipid carbons. We got stuck at the point
of running a simulation with pmemd.cuda, where we received the "num_extra"
error. Perhaps the problem here could be assigning an atomic number of -1
(not any element). What effect does the atomic number have on the
simulation -- or the parameters selected by tleap? Is there another value
we could use for the atomic number, given that we are modifying the
Lennard-Jones type before running the simulation?
If needed, Qi can provide the specific series of commands used to generate
the prmtop file with Parmed. In short, we used the outparm command. When we
tried parmout, instead, Parmed did not generate a prmtop but gave no error.
Thanks for your help.
Best,
Sally
Sally Pias
sallypias.gmail.com
On Fri, Sep 30, 2022 at 9:18 AM David A Case via AMBER <amber.ambermd.org>
wrote:
> On Thu, Sep 29, 2022, Qi Wang via AMBER wrote:
> >
> >We are trying to set up a system including dummy atoms. We got "Error in
> >numextra_test", even though the value of "NUMEXTRA" in the prmtop is the
> >same as the number of dummy particles. The prmtop file was generated with
> >parmed. Is any documentation available that could help us resolve this
> >error?
>
> Can you provide more information here:
>
> a. exactly how did you prepare the prmtop file?
> b. what commands did you use that generated the "error in
> numextra_test"? Was it a sander run?
>
> Note the the NUMEXTRA value in the prmtop file must equal the number of
> atoms with symbol "EP ". Are all your dummy atoms named "EP"?
>
> ....thx...dac
>
>
> _______________________________________________
> 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 Oct 03 2022 - 12:30:03 PDT
