In explicit solvent, the counterions are important. In principle, only
net-neutral cells have convergent energies in the Ewald summation, and
there has been work by the Brooks group IIRC, that there can be pressure
artifacts for applying a net neutralizing plasma (the approach normally
taken when running charged unit cells) in PME. While there's a reasonable
correction, I don't think it's implemented in Amber.
In any case, people use counterions in their original simulations all the
time, and then just strip them before performing MMPBSA analysis. You can
use ante-MMPBSA.py to strip out the solvent and ions from your original
topology file to generate a set of topology files compatible with
MMPBSA.py. Just make sure you strip out all of the waters and ions in both
ante-MMPBSA.py (when creating the prmtops) and MMPBSA.py (when creating the
trajectories).
HTH,
Jason
On Fri, Jun 15, 2012 at 8:25 AM, qiao xue <xueqiaoup.gmail.com> wrote:
> Thanks, you really help me a lot.
> I made new prmtop which has no Cl- ion. I think this may be comparable
> with the trajectory which has been dropped the Cl- by strip_mask. But
> the prmtop still does not match the trajectory. So I have to keep the
> Cl- in the trajectory. But this affect the energy. Maybe next time I
> would not add any countered ions at the beginning. ps: Is the
> countered ions very important to the system?
>
> On 6/15/12, Bill Miller III <brmilleriii.gmail.com> wrote:
> > Yes, the chloride ions will have an effect on the energy. In general,
> > MM-PB(GB)SA is not parameterized to include explicit ions. We generally
> > advise that users remove all ions from their trajectories (which is why
> > that is the default strip_mask for MMPBSA.py) before performing MM-PBSA
> > calculations since the PB or GB model should include ionic strength
> > effects, although obviously not in an explicit manner.
> >
> > I hope that helps.
> >
> > -Bill
> >
> > On Fri, Jun 15, 2012 at 8:01 AM, qiao xue <xueqiaoup.gmail.com> wrote:
> >
> >> Thanks, I modified the strip_mask to ":WAT" Now, the result seems
> >> good. I still have a question. Now in the system, Cl- exists. But what
> >> I want is the free energy between ligand and protein. Does the Cl-
> >> ions have any effect on the energy?
> >>
> >> On 6/14/12, Cao Yang <muxiachuixue.163.com> wrote:
> >> > The problem is caused by Cl- of your system. Just as Bill said, in
> >> default,
> >> > MMPBSA strips the Cl- and thus leading to the inconsistence between
> >> > your
> >> > prmtop file and mdcrd file. Read the strip_mask variable of &general
> >> > in
> >> > AmberTools manual and you would have some idea.
> >> >
> >> >
> >> >
> >> >
> >> >
> >> > At 2012-06-13 19:53:23,"Bill Miller III" <brmilleriii.gmail.com>
> wrote:
> >> >>Energies this high are usually indicative of an inconsistency between
> >> >> the
> >> >> prmtop and trajectory files. You should try visualizing the
> >> >> trajectories
> >> >> produced by MMPBSA.py with the corresponding prmtop files to see if
> >> >> anything looks wrong.
> >> >>
> >> >>-Bill
> >> >>
> >> >>
> >> >>
> >> >>On Jun 13, 2012, at 2:32 AM, qiao xue <xueqiaoup.gmail.com> wrote:
> >> >>
> >> >>> I do not want to include Cl-, I thought the problem may be caused by
> >> >>> Cl- in the beginning. But I have dropped Cl-, the problem still
> >> >>> exist.
> >> >>> So I do not know what this problem is.
> >> >>>
> >> >>> On 6/13/12, Bill Miller III <brmilleriii.gmail.com> wrote:
> >> >>>> MMPBSA.py, by default, removes all the ions from the original
> >> >>>> trajectory.
> >> >>>> You will need to specify the strip_mask control variable in your
> >> >>>> input
> >> >>>> file
> >> >>>> so that MMPBSA.py does not remove the chloride ion that you are
> >> >>>> attempting
> >> >>>> to include in the calculation.
> >> >>>>
> >> >>>> -Bill
> >> >>>>
> >> >>>> On Wed, Jun 13, 2012 at 2:03 AM, qiao xue <xueqiaoup.gmail.com>
> >> wrote:
> >> >>>>
> >> >>>>> Hello all,
> >> >>>>> I take a MD simulation of a protein contained Fe ion. This
> >> >>>>> protein
> >> >>>>> is a receptor. And there are several ligands. So in some of them I
> >> add
> >> >>>>> countered ions Cl-. Then I did MMPBSA calculation for these
> >> >>>>> protein-ligand complex. All the systems with countered ion Cl- is
> >> >>>>> abnormal. Like this
> >> >>>>>
> >> >>>>>
> >> >>>>> Differences (Complex - Receptor - Ligand):
> >> >>>>> Energy Component Average Std. Dev. Std.
> >> Err.
> >> >>>>> of
> >> >>>>> Mean
> >> >>>>>
> >> >>>>>
> >>
> -------------------------------------------------------------------------------
> >> >>>>> BOND 0.0000 0.0001
> >> >>>>> 0.0000
> >> >>>>> ANGLE -0.0000 0.0000
> >> >>>>> 0.0000
> >> >>>>> DIHED -1.9605 1.1967
> >> >>>>> 0.2443
> >> >>>>> VDWAALS 37918031025063.1484 125947828014390.9375
> >> >>>>> 25708992737255.8711
> >> >>>>> EEL 175.6232 121.1990
> >> >>>>> 24.7396
> >> >>>>> 1-4 VDW -831.8288 3735.5515
> >> >>>>> 762.5163
> >> >>>>> 1-4 EEL 0.0808 0.1088
> >> >>>>> 0.0222
> >> >>>>> EGB 63.4121 68.6816
> >> >>>>> 14.0196
> >> >>>>> ESURF -3.4046 1.7772
> >> >>>>> 0.3628
> >> >>>>>
> >> >>>>> DELTA G gas 37918031024810.1719 125947828014502.9375
> >> >>>>> 25708992737278.7305
> >> >>>>> DELTA G solv 60.0075 68.3297
> >> >>>>> 13.9477
> >> >>>>>
> >> >>>>>
> >> >>>>> DELTA G binding = 37918031024879.3438 +/- 125947828014482.4219
> >> >>>>> 25708992737274.5430
> >> >>>>>
> >> >>>>>
> >>
> -------------------------------------------------------------------------------------------------------------
> >> >>>>>
> >> >>>>> The VDWAALS is very huge. I am sure the parameters strip_mask is
> >> >>>>> correct. I use the default parameters. It contains Cl-. Then I
> have
> >> >>>>> tried to drop the Cl- from the trajectory. And I use the new
> prmtop
> >> >>>>> file which has no countered ion. But the result is still wrong.
> >> >>>>>
> >> >>>>> But the system without countered ion Cl- is good. The result is
> >> >>>>> this:
> >> >>>>>
> >> >>>>> Differences (Complex - Receptor - Ligand):
> >> >>>>> Energy Component Average Std. Dev. Std.
> >> Err.
> >> >>>>> of
> >> >>>>> Mean
> >> >>>>>
> >> >>>>>
> >>
> -------------------------------------------------------------------------------
> >> >>>>> BOND -0.0000 0.0000
> >> >>>>> 0.0000
> >> >>>>> ANGLE 0.0000 0.0000
> >> >>>>> 0.0000
> >> >>>>> DIHED -0.0676 0.0682
> >> >>>>> 0.0014
> >> >>>>> VDWAALS -56.3502 3.2913
> >> >>>>> 0.0658
> >> >>>>> EEL -20.1411 4.8748
> >> >>>>> 0.0975
> >> >>>>> 1-4 VDW 3.0567 1.1338
> >> >>>>> 0.0227
> >> >>>>> 1-4 EEL 0.1026 0.0240
> >> >>>>> 0.0005
> >> >>>>> EPB 50.0003 4.5535
> >> >>>>> 0.0911
> >> >>>>> ECAVITY -4.8209 0.1075
> >> >>>>> 0.0021
> >> >>>>>
> >> >>>>> DELTA G gas -73.3996 5.4641
> >> >>>>> 0.1093
> >> >>>>> DELTA G solv 45.1794 4.5587
> >> >>>>> 0.0912
> >> >>>>>
> >> >>>>>
> >> >>>>> DELTA G binding = -28.2202 +/- 4.8647
> >> >>>>> 0.0973
> >> >>>>>
> >> >>>>>
> >> >>>>>
> >>
> ------------------------------------------------------------------------------------------------------------------
> >> >>>>>
> >> >>>>> This VDWAALS seems good.
> >> >>>>>
> >> >>>>>
> >> >>>>> This input file is same:
> >> >>>>> &general
> >> >>>>> endframe=2500, keep_files=2,
> >> >>>>> /
> >> >>>>> &gb
> >> >>>>> igb=2, saltcon=0.100,
> >> >>>>> /
> >> >>>>> &pb
> >> >>>>> istrng=0.100,
> >> >>>>> /
> >> >>>>>
> >> >>>>> -----------------------------------------------------------
> >> >>>>>
> >> >>>>> The two systems is the same protein. The ligands are very normal.
> >> >>>>> The
> >> >>>>> most different thing is the countered ion. I think this may be
> >> >>>>> caused
> >> >>>>> by Cl- at first. But I have droped this ion. I do not know why
> this
> >> is
> >> >>>>> still wrong. I hope someone could help me.
> >> >>>>> Thank you very much!
> >> >>>>>
> >> >>>>>
> >> >>>>> Xue Qiao.
> >> >>>>>
> >> >>>>> _______________________________________________
> >> >>>>> AMBER mailing list
> >> >>>>> AMBER.ambermd.org
> >> >>>>> http://lists.ambermd.org/mailman/listinfo/amber
> >> >>>>>
> >> >>>>
> >> >>>>
> >> >>>>
> >> >>>> --
> >> >>>> Bill Miller III
> >> >>>> Quantum Theory Project,
> >> >>>> University of Florida
> >> >>>> Ph.D. Graduate Student
> >> >>>> 352-392-6715
> >> >>>> _______________________________________________
> >> >>>> AMBER mailing list
> >> >>>> AMBER.ambermd.org
> >> >>>> http://lists.ambermd.org/mailman/listinfo/amber
> >> >>>>
> >> >>>
> >> >>> _______________________________________________
> >> >>> AMBER mailing list
> >> >>> AMBER.ambermd.org
> >> >>> http://lists.ambermd.org/mailman/listinfo/amber
> >> >>
> >> >>_______________________________________________
> >> >>AMBER mailing list
> >> >>AMBER.ambermd.org
> >> >>http://lists.ambermd.org/mailman/listinfo/amber
> >> > _______________________________________________
> >> > AMBER mailing list
> >> > AMBER.ambermd.org
> >> > http://lists.ambermd.org/mailman/listinfo/amber
> >> >
> >>
> >> _______________________________________________
> >> AMBER mailing list
> >> AMBER.ambermd.org
> >> http://lists.ambermd.org/mailman/listinfo/amber
> >>
> >
> >
> >
> > --
> > Bill Miller III
> > Quantum Theory Project,
> > University of Florida
> > Ph.D. Graduate Student
> > 352-392-6715
> > _______________________________________________
> > AMBER mailing list
> > AMBER.ambermd.org
> > http://lists.ambermd.org/mailman/listinfo/amber
> >
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
--
Jason M. Swails
Quantum Theory Project,
University of Florida
Ph.D. Candidate
352-392-4032
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Jun 15 2012 - 06:30:02 PDT
