AMBER Archive (2008)

Subject: Re: AMBER: Chloride anion ligands leave the protein during MD

From: Jiri Sponer (sponer_at_ncbr.chemi.muni.cz)
Date: Fri Nov 14 2008 - 08:19:33 CST


There is one another problem with Cl-.
As anions, they have electron densities reaching far from the
atomic centers and thus are highly polarisable.
In principle, accurate description by pair additive force
fields is not achievable, though due to compensation
of errors part of the deficiencies can be eliminated.
Nevertheless, even if everything perfect (solvation, concentration)
the results could, although not always, be biased.

Best wishes Jiri

-------------------------------------------------------
Jiri Sponer
Institute of Biophysics
Academy of Sciences of the Czech Republic
Kralovopolska 135
CZ-61265 Brno
Czech Republic
e-mail: sponer_at_ncbr.chemi.muni.cz
fax: 420 5412 12179
phone: 420 5415 17133
http://www.ibp.cz/labs/LSDNA/
http://www.springer.com/east/home/chemistry?SGWID=5-135-22-158967829-0
-----------------------------------------------------------

[ Charset ISO-8859-1 unsupported, converting... ]
> I have a substantially more basic question.
>
> Why would you expect a chloride anion to remain attach to the protein in
> implicit solvent ?
>
> I could potentially understand in explicit solvation, but in GB ? There
> is basically NOTHING outside the protein that would keep it where it
> belongs. Fluctuations will make its position change, but once it is out,
> it is out, and of course it would fly away.
>
> Experimentally, what people might see is ONE chloride anion in a place,
> but of course it is not always the SAME chloride anion. In your
> simulations, once the chloride anion leaves, there is no ionic
> concentration to replace it.
>
> Adrian
>
>
> Gustavo Seabra wrote:
> > Hi Francesco,
> >
> > I'm sorry i got a bit confused with all the inputs... At which point
> > exactly does the Cl- ion start to drift? Is it as soon as you start
> > MD, or only when you add nrespa=2? (none of your inputs have this).
> >
> > Gustavo.
> >
> > On Thu, Nov 13, 2008 at 10:06 AM, Francesco Pietra
> > <chiendarret_at_gmail.com> wrote:
> >> Hi:
> >> I have carried out docking of a small protein against a large one, the
> >> latter generated with Modeller from an extremely similar natural
> >> protein and carrying chloride anion ligands. These were accounted for
> >> during docking, and in fact they conserved their position after
> >> docking. The docking region does not involve the chloride anions
> >> directly.
> >>
> >> With Amber 10 and pmemd, the complex was first minimized under GB conditions:
> >>
> >> &cntrl
> >> imin=1,
> >> maxcyc=3000,
> >> ncyc=1500,
> >> ntb=0,
> >> igb=5,
> >> cut=999
> >> /
> >>
> >>
> >> Then it was heated gradually to 300K and further equilibrated as follows::
> >>
> >> &cntrl
> >> imin=0, irest=0, ntx=1, ntb=0,
> >> igb=5, ntc=2, ntf=2,
> >> ntt=3, gamma_ln=2.0,
> >> nstlim=2500, dt=0.001,
> >> ntpr=100, ntwx=100,
> >> tempi=100.0, temp0=150.0,
> >> cut=999.0, rgbmax=999.0,
> >> nmropt=1
> >> /
> >> &wt TYPE='TEMP0', istep1=0, istep2=2500,
> >> value1=100.0, value2=150.0,
> >> /
> >> &wt TYPE='END'
> >> /
> >>
> >> &cntrl
> >> imin=0, irest=0, ntx=1, ntb=0,
> >> igb=5, ntc=2, ntf=2,
> >> ntt=3, gamma_ln=2.0,
> >> nstlim=2500, dt=0.001,
> >> ntpr=100, ntwx=100,
> >> tempi=150.0, temp0=200.0,
> >> cut=999.0, rgbmax=999.0,
> >> nmropt=1
> >> /
> >> &wt TYPE='TEMP0', istep1=0, istep2=2500,
> >> value1=150.0, value2=200.0,
> >> /
> >> &wt TYPE='END'
> >> /
> >>
> >> &cntrl
> >> imin=0, irest=0, ntx=1, ntb=0,
> >> igb=5, ntc=2, ntf=2,
> >> ntt=3, gamma_ln=2.0,
> >> nstlim=2500, dt=0.001,
> >> ntpr=100, ntwx=100,
> >> tempi=200.0, temp0=250.0,
> >> cut=999.0, rgbmax=999.0,
> >> nmropt=1
> >> /
> >> &wt TYPE='TEMP0', istep1=0, istep2=2500,
> >> value1=200.0, value2=250.0,
> >> /
> >> &wt TYPE='END'
> >> /
> >>
> >> &cntrl
> >> imin=0, irest=0, ntx=1, ntb=0,
> >> igb=5, ntc=2, ntf=2,
> >> ntt=3, gamma_ln=2.0,
> >> nstlim=2500, dt=0.001,
> >> ntpr=100, ntwx=100,
> >> tempi=250.0, temp0=300.0,
> >> cut=999.0, rgbmax=999.0,
> >> nmropt=1
> >> /
> >> &wt TYPE='TEMP0', istep1=0, istep2=2500,
> >> value1=250.0, value2=300.0,
> >> /
> >> &wt TYPE='END'
> >> /
> >>
> >> Then increased dt!!
> >>
> >> &cntrl
> >> imin=0, irest=0, ntx=1, ntb=0,
> >> igb=5, ntc=2, ntf=2,
> >> ntt=3, gamma_ln=2.0,
> >> nstlim=3000, dt=0.002,
> >> ntpr=100, ntwx=100,
> >> tempi=290.0, temp0=300.0,
> >> cut=999.0, rgbmax=999.0,
> >> nmropt=1
> >> /
> >> &wt TYPE='TEMP0', istep1=0, istep2=600,
> >> value1=290.0, value2=300.0,
> >> /
> >> &wt TYPE='END'
> >> /
> >>
> >>
> >>
> >> As EPTOT ETOT EKTOT were very nearly constant and the whole complex
> >> had not suffered visible distotion, I tried MD of this huge,
> >> unabridged complex under GB conditions in very small chunks as
> >>
> >> Production under GB conditions
> >> &cntrl
> >> imin=0, irest=1, ntx=5, ntb=0,
> >> igb=5, ntc=2, ntf=2,
> >> ntt=3, gamma_ln=2.0,
> >> nstlim=3000, dt=0.002,
> >> ntpr=100, ntwx=100,
> >> tempi=300.0, temp0=300.0,
> >> cut=999.0, rgbmax=999.0
> >> /
> >>
> >>
> >> Starting from production 2, I added nrespa=2 (reaching the guard limit
> >> of 4femtosec)
> >>
> >> Surprisingly (for me) the chloride anions started to leave the
> >> protein, while no other distortion showed up. While at frame 0 the
> >> chloride anions are their positions, at frame 50(a base some 25 ps of
> >> trajectory) they are out of the proteins a gradually they disappear
> >> from the screen. Evaluating rmsd with ref frame 0, rmsd increases
> >> linearly with frame number, and in the attached plot. The minimum
> >> potential energy remains at frame 100, out of 250 frames, and ETOT
> >> (black), EPTOT (green), and EKTOT (red) seem OK as in the attached
> >> plot. Nonetheless, might the large nrespa=2 be responsible for the
> >> instability of chloride anion ligands? The size of the system is such
> >> that the 50% gain with nrespa=2 is difficult to renounce to.
> >>
> >> Chloride anion was set as Cl- and xleap did not complain about.
> >>
> >> Sorry if I made a long story for something trivial, obvious to
> >> experienced people.
> >>
> >> Thanks for any kind of comment
> >> francesco pietra
> > -----------------------------------------------------------------------
> > The AMBER Mail Reflector
> > To post, send mail to amber_at_scripps.edu
> > To unsubscribe, send "unsubscribe amber" (in the *body* of the email)
> > to majordomo_at_scripps.edu
> >
>
> --
> Dr. Adrian E. Roitberg
> Associate Professor
> Quantum Theory Project
> Department of Chemistry
>
> Senior Editor. Journal of Physical Chemistry
> American Chemical Society
>
> University of Florida PHONE 352 392-6972
> P.O. Box 118435 FAX 352 392-8722
> Gainesville, FL 32611-8435 Email adrian_at_qtp.ufl.edu
> -----------------------------------------------------------------------
> The AMBER Mail Reflector
> To post, send mail to amber_at_scripps.edu
> To unsubscribe, send "unsubscribe amber" (in the *body* of the email)
> to majordomo_at_scripps.edu
>
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" (in the *body* of the email)
      to majordomo_at_scripps.edu