On Wed, 2014-04-16 at 12:11 +0800, Jinfeng Huang wrote:
> Thanks for your helps! The explanation is very clear.
> My amber version is Amber 11. The input file I used was:
> -------- input for WAT energy calculation--------------
> &cntrl
> imin=1,
> maxcyc=0,
> ncyc=0,
> ntb=0,
> cut=9999.0
> /
> -------------------------------------------------------
> By playing around the water molecule to further understand the
> "excluded_nonbond_list", I end up to have one more question. What I
> did is:
> 1). generate prep and frcmod files for a single water molecule with
> H-H disconnected. The residue name is set as "WT1" to differ from TP3
> model.
> 2). atom types for oxygen and hydrogens are OW and HW, respectively.
> Then, sander calculates nonbond electrostatics with the above sander
> controlling input file
> 3). atom types for oxygen and hydrogens are OW and HO, respectively,
> Then, sander gives zero for nonbond electrostatics.
>
> It looks like whenever there is HW atom, the nonbond electrostatics
> are calculated in PME code. Does it mean the PME code ignores the
> angle term for HW atom type?
To my knowledge, the OW, and HW atom types are only used to identify
waters whose constraints can be computed analytically via SETTLE.
Are you sure that the angle parameters exist in all of the topology
files? You should use the "printBonds" or "printAngles" command in
ParmEd to make sure that it contains the bonds and angles that you
expect. You're trying things here that most other people don't, so we
may not know the answer to why certain things happen when you break
assumptions that sander and/or pmemd make about your system.
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 Wed Apr 16 2014 - 08:30:03 PDT
