AMBER Archive (2005)

Subject: Re: AMBER: Questions about sander.QMMM

From: M. L. Dodson (bdodson_at_scms.utmb.edu)
Date: Fri Jan 07 2005 - 10:45:45 CST

Apologize to the list for following up on my own post.

On Friday 07 January 2005 10:26 am, M. L. Dodson wrote:
> On Friday 07 January 2005 09:51 am, Cenk Andac wrote:
> > Dear amber community,
> >
> > I will have two questions regarding
> > sander.QMMM
> >
> > 1) I was wondering if it is possible to run
> > sander.QMMM using NOE restraints.
> >
> > I added a solvatecap around the center of my molecule
> > in xleap as follows
> >
> > >solvatecap MOL TIP3PBOX MOL.MOL.O1 15 0.7
> > >Center of solvent box is:
> > -5.456166, -10.706805, -8.113191
> > >Added 400 residues.
> > >saveamberparm MOL MOL.prmtop MOL.x
> >
> > I checked my unit in xleap and it gave no errors.
> >
> > Then I relaxed my MOL using sander.QMMM with the following
> > divcon.in and relax.in parameters:
> >
> > divcon.in :
> > CARTESIAN PM3 CHARGE=4 &
> > STANDARD DIRECT
> > END_COORD
> >
> > relax.in:
> > &cntrl
> > imin=1, maxcyc=500, ncyc=250,
> > cut=20, ntb=0, fcap=1.5,
> > ifqt=1, nqt=52, idc=0,
> > /
> > 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
> > 16 17 18 19 20 21 22 23 24 25 26 27
> > 28 29 30 31 32 33 34 35 36 37 38 39
> > 40 41 42 43 44 45 46 47 48 49 50 51
> > 52
> > END
> > END
> >
> > which gave no errors during relaxation.
> >
> > However, I encountered a problem when
> > I tried to run sander.QMMD with the following
> > input parameter which requires use of NOE
> > restarints. Of course, I carried out the second
> > sander.QMMM run for relaxed coordinates.
> >
> > qmmm.in:
> >
> > 300K constant temp QMMM
> > &cntrl
> > imin=0, ntb=0, pencut=-0.001,
> > cut=20, fcap=1.5,
> > tempi=0, temp0=300.0,
> > ntt=3, gamma_ln=10, vlimit=10,
> > nstlim=1000, dt=0.001,
> > ntpr=10, ntwx=10, nmropt=1,
> > ifqt=1, nqt=12, idc=0,
> > /
> > 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
> > 16 17 18 19 20 21 22 23 24 25 26 27
> > 28 29 30 31 32 33 34 35 36 37 38 39
> > 40 41 42 43 44 45 46 47 48 49 50 51
> > 52
> > /
> > &wt type='REST',
> > istep1=0,istep2=500,
> > value1=0.1, value2=1.0, /
> > &wt type='REST',
> > istep1=501,istep2=1000,
> > value1=1.0, value2=1.0, /
> > &wt type='END' /
> > LISTOUT=POUT
> > DISANG=RST33gg-2
> > END
> > END
> >
> > Consequently I received the following error message
> > in qmmm.out file
> >
> > Number of triangulated 3-point waters found: 400
> > FATAL ERROR
> > Link atoms MUST be optimized before MD run
> >
> >
> > Could anyone please let me know how to optimize the link atoms
> > or let me know if there is anything wrong with my input
> > parameters?
> >
>
> The restart file for a coupled potential run is (very slightly)
> different from an ordinary MD restart file (has to have
> coordinates for the link atoms in it). You need to do a coupled
> potential MINIMIZATION (1 cycle is sufficient to satisfy the
> software, and may even be sufficient in other aspects if the
> structure is already well relaxed with respect to the FF) to
> generate an appropriate restart file. Then use the restart file
> from the minimization run as the starting structure for the
> coupled potential MD run.

I see your relax step is a minimization.  Sorry.  Are you sure you
used the restart file from the minimization as the starting
structure for the MD run?

Bud Dodson
>
> > 2)Is it a good idea to incorporate NOE restraints during QMMM
> > runs for solvated systems? Would it be a better idea to NMR
> > restrain my molecule in implicit water first, then take the
> > final structure of my molecule, solvate it and run sander.QMMM
> > putting a restrain for the final structure?
> >
> > 3) Is it possible to score final QMMM structures using
> > mm_pbsa? As far as I understand from the QMMM tutorial (2004),
> > a solvatebox is not added in prmtop files. I have tried running
> > mm_pbsa for an QMMM coordinate output, however, mm_pbsa was
> > unable to execute pbsa. Do I need to define a solvatebox in my
> > prmtop file while running sander.QMMM so that mm_pbsa will be ok
> > with sander.QMMM output.
> >
> > cheers,
> >
> > jenk.
> >
> >
>
> It might be better to describe to the list what you are trying to
> accomplish and get knowledgeable comments. Coupled potential runs
> are quite "tricky" to setup and interpret, in my experience.
>
> Bud Dodson
> 

-- 
M. L. Dodson                                bdodson_at_scms.utmb.edu
409-772-2178                                FAX: 409-747-8608

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