On Fri, 2015-07-10 at 16:53 +0800, fjliu wrote:
> Dear all:
> I want to imitate the tutorial A2 to do the QM/MM simulation
> with the water model replaced by amoeba. Then I do as follows:
> leapin
> xleap -s -f
> NMA = loadpdb NMA_skeleton.pdb
> edit NMA (do the relax)
> solvatebox NMA TIP3PBOX 15
> savepdb NMA NMA_slv.pdb
> saveamberparm NMA NMA_slv.top NMA_slv.crd(save to see the box
> information)
> quit
> Then:
> pdbxyz NMA_slv
> the NMA_slv.key:
> parameters /home/fjliu/software/tinker/params/amoebabio09
> a-axis 40.18
> b-axis 39.30
> c-axis 39.56
>
> I get the NMA_slv.xyz. Followed by: analyze NMA_slv PC >
> NMA_slv.analout, I got the error message:
>
> Undefined Torsional Parameters :
>
> Type Atom Names Atom
> Classes
>
> Torsion 1-C 2-C 7-N 8-C 40 3
> 1 40
>
> MECHANIC -- Some Required Potential Energy Parameters are
> Undefined
>
> TINKER is Unable to Continue; Terminating the Current Calculation
>
> What is wrong with it?
As the error message states, the Tinker parameter set does not define
all of the torsions defined in your molecule. In particular, the atom
types C-C-N-C has no torsion defined. So you need to add a definition
for this torsion to your key file so Tinker knows how to assign it.
What I did to parametrize NMA was to look through the amoebabio09
parameter set, look for a torsion parameter whose atom types are as
close as possible to the ones in NMA, and copy those parameters to my
key file. Ultimately, however, this is a question about Tinker, not
really Amber, so we may not be the most expert authorities on what went
wrong.
> Also I do the example in
> /home/fjliu/software/amber14/src/pmemd.amoeba/build_amoeba/examples/JAC,thre
> is no problem with the given jac.pdb. But if I extracted a water
> molecule to do the pdbxyz, the atom type information will not give out,
> then it will be wrong when I do the analyze.
>
> Another question is can I do the QM/MM calculation with amoeba
> water model?
Not with Amber.
HTH,
Jason
--
Jason M. Swails
BioMaPS,
Rutgers University
Postdoctoral Researcher
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Jul 10 2015 - 05:00:02 PDT
