Hi Prof. Luo,
Thanks for the clarification.
I think EEL energies should be scaled by the interior dielectric constant.
I was curious why there isn't a corresponding variable to do it in
MMPBSA.py. But yes, we can do it by hand.
Best,
Qiong
On Wed, Jul 25, 2012 at 12:44 PM, Ray Luo, Ph.D. <ray.luo.uci.edu> wrote:
> Unfortunately, the current script doesn't support a non-unity interior
> dielectric constant. However, this is quite straightforward to do as you
> pointed from the cited paper in your email. You simple divide all the
> intramolecular EEL energies by the interior dielectric constant by hand.
> The scaled EEL energies are then consistent with the EPB energies computed
> with the same interior dielectric constant.
>
> This also comes to the Dave's original suggestion of not treating any
> automatic script as a black box. Try to understand each step of the
> calculations and make sure you can do these calculations by hand. If you
> fully understand the script, it's quite straightforward to tailor the
> output to suit your specific project at hand.
>
> All the best,
> Ray
>
> On Wed, Jul 25, 2012 at 7:36 AM, Qiong Zhang <qiong.zhang.qzh.gmail.com>wrote:
>
>> Hi Jason,
>> Thanks for the reply.
>> I read the paper on J. Chem. Inf. Model. 2011, 51, 69–82. It shows how
>> dielectric constant affects the binding free energy calculated by MMPBSA.
>> When I looked closely at their data (Table S2) the gas-phase electrostatic
>> interactions (EEL) are also scaled by solute dielectric constant.
>> Indi=1
>> deltaEEL deltaEPB
>> -133.38±1.13 193.46±0.23
>> Indi=2
>> deltaEEL deltaEPB
>> -66.69±0.56 93.87±0.16
>>
>> I found in this thread http://archive.ambermd.org/201106/0170.html,
>> Prof. Luo has commented that:
>> "In general, a higher dielectric constant should reduce the contribution
>> of electrostatic interactions (both Coulombic and solvation electrostatic,
>> i.e. PB/GB) in the final binding free energies. "
>>
>> "Note that the division of what constitutes 'EPB' and 'EEL' is rather
>> arbitrary and artificial (they are calculated together and decomposed
>> after-the-fact, I think)".
>> I thought they are calculated separately. As far as I understood, for
>> "EEL" part, it is calculated directly by Coulombic law; while for 'EPB' is
>> calculated by solving the PB equations. So for "EEL" part, it should be
>> divided by the dielectric constant when solute dielectric constant is not
>> equal 1. Please correct me if I am wrong here.
>>
>> Whether gas-phase electrostatic interactions should be scaled by the
>> solute dielectric constant or not if one would like to see how the
>> dielectric constant affects the binding free energy?
>>
>> Please help to shed more lights on it. Thanks!
>> Best,
>> Qiong
>>
>>
>>
>> On Wed, Jul 25, 2012 at 10:11 AM, Jason Swails <jason.swails.gmail.com>wrote:
>>
>>> On Wed, Jul 25, 2012 at 9:15 AM, Qiong Zhang <qiong.zhang.qzh.gmail.com
>>> >wrote:
>>>
>>> > I am using Amber11/AT1.5. Could anyone point to me how can I get
>>> > electrostatic energies in gas phase scaled by dielectric constant with
>>> > MMPBSA.py or which variable should I use please? I've read the related
>>> part
>>> > for MMPBSA.py in the manual, but have not found it.
>>> > With the following input, only the polar solvation energy is scaled,
>>> while
>>> > the electrostatic energies in gas phase stays unaffected.
>>> >
>>>
>>> This is expected behavior, as far as I understand. The effect of the
>>> internal dielectric constant is reflected in the polar solvation term
>>> (EPB), not the gas-phase electrostatic interactions (EEL), but this
>>> doesn't
>>> change the nature of that difference.
>>>
>>> Note that the division of what constitutes 'EPB' and 'EEL' is rather
>>> arbitrary and artificial (they are calculated together and decomposed
>>> after-the-fact, I think), and terms involving the dielectric constant
>>> (either the internal or external) are rolled into EPB.
>>>
>>> HTH,
>>> Jason
>>>
>>> --
>>> 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
>>>
>>
>>
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed Jul 25 2012 - 10:00:04 PDT
