The force field you load will be applied to your "ligand" molecules,
including all water molecules in the simulation. If you feel that liganded
water molecules have different characteristics than water molecules far
away from the protein ("in the bulk"), this is an age-old debate about
whether explicit accounting for polarization is needed. You are definitely
on the right track, but even in the case you are looking at the "liganded"
water molecule will probably exchange many times with solvent waters, and
if the electron distribution or precise molecular shape of the ligand water
change upon binding, then the newly liganded water will take on those
characteristics and the formerly liganded water will return to its "normal"
state in the bulk. Your simulation must anticipate that the liganded water
molecule will exchange in the same way, so without full treatment of
polarization (Amoeba, or one of the CHARMM community's SWM4 models, for
instance) all of the waters must be the same and a water model that gets
the right bulk properties is the appropriate choice.
Now, as to the particulars of how you might proceed:
- Use ff14SB, no ff12SB as Carlos said (12 was a milestone on the evolution
of 14)
- If you don't like TIP3P, no one will really hold a grudge if you instead
use SPC/E with ff14SB.
- In order to use SPC/E, you will need to source leaprc.water.spce in your
tleap input AFTER sourcing ff14SB
Note that sourcing leaprc.water.spce will also load a different frcmod file
containing ion parameters optimized for SPC/E water. Again, this applies
to interactions between the ions and bulk water, and is not necessarily the
best for the bound ions your system contains. However, if you've got to
choose one way to model the ions, their characteristics in bulk water are
the way to go.
Good luck!
Dave
On Mon, Apr 11, 2016 at 10:02 AM, wjli <wenjuanli93.gmail.com> wrote:
> If I source leaprc.ff12SB,ions08.lib and solvents.lib will be loaded.
> Whether it's okay to apply them to my ligands(Na+,Cl-,H2O) when I want to
> calculate PMF.
>
>
>
>
> ------------------ Original ------------------
> From: "carlos.simmerling";<carlos.simmerling.gmail.com>;
> Date: Mon, Apr 11, 2016 09:50 PM
> To: "AMBER Mailing List"<amber.ambermd.org>;
>
> Subject: Re: [AMBER] ff12SB force field
>
>
>
> I don't really understand your question. ff12SB is for proteins, and is not
> related to the "(Cl- or Na+ or H2O)" that you mention. If you are
> simulating proteins, we suggest the newer ff14SB. If you have a question
> about the force field, can you try to be more specific about your concern?
>
> On Mon, Apr 11, 2016 at 9:47 AM, wjli <wenjuanli93.gmail.com> wrote:
>
> > Dear Amber members,
> >
> > I want to calculate PMF profiles for ligands(Cl- or Na+ or H2O) via
> tunnel
> > . I hesitate about the property of using ff12SB force field,just like
> > applying water box's force field(TIP3PBOX) to ligand water.
> >
> > I will very appreciate if anyone can offer some suggestions.
> >
> > Look forward to your reply!
> > Best wishes.
> > wjli
> > _______________________________________________
> > 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
Received on Mon Apr 11 2016 - 14:00:03 PDT
