AMBER Archive (2006)

Subject: Re: AMBER: acidifying a water box

From: Thomas Cheatham (tec3_at_utah.edu)
Date: Thu Sep 14 2006 - 17:56:32 CDT


> I'm attempting to model a water/D2O system (298 molecules each) with varying
> levels of HCl. I'm not sure exactly how to pursue this in AMBER (7.0), but
> I've taken the following approach which causes sander to crash during a
> minimization:

All of your building procedure seems fine, however without the
prmtop/inpcrd files it is hard to decipher further.

> 2) A script then handles population of the (26 angstrom cubic) system
> (attached - out1.pdb)

Of key concern is step (2) above. How is this generated and where are the
ions placed? From the minimization output, initially you have extremely
high VDW (which implies overlap) which then relaxes to overly stretch
bonds and likely has overlapping atoms (oxygens) which lack vdw radii so
you get a charge/polarization catastrophe.

> NSTEP ENERGY RMS GMAX NAME NUMBER
> 1 NAN NAN 0.0000E+00 O 1
> BOND = 0.0559 ANGLE = 0.0002 DIHED = 0.0000
> VDWAALS = nan EEL = nan HBOND =
> EPOLAR = nan ETHREEB = 0.0000

All energy is in the VDW/POLARIZATION/ELECTROSTATICS => overlap

You can look for overlap by using ptraj and the "checkoverlap" command.

ptraj prmtop << EOF
trajin inpcrd
checkoverlap
EOF

Resolving this is going to require either (1) building a better initial
set of coordinates, (2) using LEaP to build it for you, or (3) careful
runtime trickery to get it to relax properly.

(1) Build better initial script/PDB file; it is curious when I view your
PDB file that it appears (with my viewer) to look like a 9x9x9 water box
with an extra set of 18 waters on one side. If these overlap when imaged,
you are toast. Make sure that your intial ion positions and waters do not
overlap.

(2) Use LEaP to build the box. I've just done this recently for a related
system, specifically Na+, Cl- in water. What I do is add a single ion,
solvate, and then use the addions command sequentially to add more ions.
(I do it sequentially to avoid ion overlap possible if it is done in a
single command). You can do this to build the initial PDB file and then
edit this file to convert it to your ion and water names, load that back
into leap with your new parameters, use the setbox command to set the box
information and off you go...

aa = sequence { Na+ }
solvateoct aa TIP3PBOX 15.0 iso
addions aa Cl- 1
addions aa Na+ 1
addions aa Cl- 1
addions aa Na+ 1
addions aa Cl- 1
addions aa Na+ 1
addions aa Cl- 1
addions aa Na+ 1
addions aa Cl- 1
addions aa Na+ 1
addions aa Cl- 1
addions aa Na+ 1
addions aa Cl- 1

(3) trickery: play with the NMR options to turn down / off electrostatics
and minimize to fix bad contacts or since your box is so ordered to begin
with anyways, use the "scale" command in ptraj to scale up the coordinates
and then equilibrate at high temperature and high pressure (500K, 500
atmospheres) to get it to settle down post minimization...

ptraj prmtop << EOF
trajin inpcrd
trajout inpcrd.mod restart
scale x 1.4 y 1.4 z 1.4 *
EOF

--tom

\-/ Thomas E. Cheatham, III (Assistant Professor) College of Pharmacy
-/- Departments of Med. Chem. and of Pharmaceutics and Pharm. Chem.
/-\ Adjunct Asst Prof of Bioeng.; Center for High Performance Computing
\-/ University of Utah, 30 S 2000 E, SH 201, Salt Lake City, UT 84112
-/-
/-\ tec3_at_utah.edu (801) 587-9652; FAX: (801) 585-9119
\-/ BPRP295A http://www.chpc.utah.edu/~cheatham

-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu