AMBER Archive (2007)

Subject: Re: AMBER: solvateCap minin and mdin for qm-mm dftb

From: M. L. Dodson (mldodson_at_houston.rr.com)
Date: Fri Aug 24 2007 - 08:11:06 CDT


Francesco Pietra wrote:
> I am trying to build prmtop and inpcrd for qm-mm with theory=7 (dftb). I am
> stuck at solvateCap, argument #3 (position).
>
> Using *.prepin and *frcmod from previous successful runs with theory=2,
>
> startx
>
> $AMBERHOME/exe/xleap -s -f $AMBERHOME/dat/leap/cmod/leaprc.ff99
>
> source leaprc.gaff
>
> loadamberprep EQE.prepin
>
> loadamberparams EQE.frcmod
>
> check EQE (unit is OK)
>
> solvateCap EQE TIP3PBOX EQE 20.0
>
> returns
>
> Argument #3 is type Unit must be of type: [molecule residue atom list]
>
> Using as Argument #3: "1" (residue number in prepin), "C17" (a central carbon
> atom number in prepin, "c" the C17 atom type, or, finally (and blindly)
> EQE.2.CA the error is: Argument #3 is type String.
>
> I am short of imagination (even consulting the manual and the mailing list) of
> what should be given in my case as Argument #3.

Without getting into the structured design of LEaP, perhaps the
easiest way to for you to proceed is to pick the atom you want to
make the center of your spherical cap. For purposes of
illustration, let us assume the Cartesian coordinates of that
atom are:

x = 1.0, y = 2.0, z = 3.0

Then a solvate cap command using these coordinates might be:

solvateCap EQE TIP3PBOX {1.0 2.0 3.0} 20.0

The desc command is your friend when trying to learn the syntax of
LEaP. You just have to practice.

> ________________--
> That said, on summing up info from the manual and suggestions from Bud and
> Gustavo, I have a number of doubts about min.in and md.in, in particular if
> qm_pme should be omitted at all. The two files I have set up read:
>
> Initial min for qmmm with dftb
> &cntrl
> imin=1, maxcyc=500, ncyc=200,
> cut=20.0, ntb=1, ntc=2, ntf=2,
> ivcap=0, fcap=10
> ifqnt=1
> /
> &qmmm
> qmmask=':1',
> qmcharge=0,
> qmtheory=7,
> qmshake=0,
> qm_ewald=0
> /
>
> 300K constant temp for qmmd with dftb
> &cntrl
> imin=0, ntb=1
> cut=20.0, ntc=2, ntf=2,
> tempi=300.0, temp0=300.0,
> ntt=3, gamma_ln=1.0,
> nstlim=20000, dt=0.002,
> ivcap=0, fcap=10,
> ntpr=100, ntwx=100,ifqnt=1
> /
> &qmmm
> qmmask=':1',
> qmcharge=0,
> qmtheory=7,
> qmshake=0,
> qm_ewald=0
> /
>
>
> I want to try if, at the current implementation of qm-mmm/dftb
> in Amber9, speed is acceptable for a 98-atoms molecule. If
> acceptable, to compare ESCF and EPtot with AM1.
>

Since amber9 sander cannot do pme with QMMM and qmtheory=7, you
should not turn on qm_pme. Also, for solvent cap simulations,
ntb=0, not ntb=1 as you have specified. And you should think
about the suggestion on page 99 of the manual to set igb=10. I
also think you may have to reduce the timestep. I use 0.5fs for
QMMM steered MD, but you may be able to get by with 1.0fs, 2.0fs
is generally workable, but is said to be borderline for classical
MD. See the email list archives.

Of course the definition of acceptable speed is personal, but I
have found 90+ atom QM regions to be workable with qmtheory=7.

>
>
> Thanks for checking
>
> francesco pietra

HTH,

Bud Dodson

-- 
M. L. Dodson
Email:	mldodson-at-houston-dot-rr-dot-com
Phone:	eight_three_two-five_63-386_one
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu