On Mon, Sep 27, 2010, Pete Kekenes-Huskey wrote:
>
> I did just try recreating the parm/crd files using leap WITHOUT the
> RNA FF and obtained the same result. I will try your other suggestions
> to see if this improves things any. I did notice that the VDW
> potential is whacky in my simulation:
>
> > sander -i min.inp -p input.parm -c input.crd -r restrt -ref input.crd
>
> NSTEP ENERGY RMS GMAX NAME NUMBER
> 10 1.2274E+10 6.6912E+08 1.7085E+11 C3' 1945
>
> BOND = 92593.0046 ANGLE = 6364.2650 DIHED = 1729.5378
> VDWAALS = ************* EEL = -224802.7617 HBOND = 0.0000
Sounds like your starting coordinates have two atoms on top of each other.
Plus terrible bond energies, etc. When I used your starting pdb file and
the other leap commands you reported, I get the following initial energies:
NSTEP ENERGY RMS GMAX NAME NUMBER
1 -1.9313E+04 1.7433E+01 6.8320E+02 H6 49
BOND = 6.6333 ANGLE = 160.3167 DIHED = 84.6469
VDWAALS = 3702.6766 EEL = -22811.2185 HBOND = 0.0000
1-4 VDW = 76.4477 1-4 EEL = -532.0970 RESTRAINT = 0.0415
EAMBER = -19312.5942
So, something bad has happened to your "input.crd" file before it ever gets
to sander. Have you run ambpdb on input.crd and looked at that pdb file in
VMD?
[It's also quite odd that atom 1945 is a C3'; the "input_sml.pdb" file you
posted only has 85 atoms. So, I'm wondering if the system you are having
problems with is really the same system that you posted the files for? For
example, my prmtop file shows 8565 atoms, but only 85 of those are RNA, four
are Na+ ions, and the rest are all water.]
....dac
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Sep 27 2010 - 09:00:04 PDT
