Dear Jason,
Thanks a lot for your reply. I didn't expect that the fluctuating charges
would have an impact on the density.
Basically, I want to run pH-REMD simulations on a protein that has several
ionisable residues interacting with one another and based on what I have
observed during the NPT equilibration, the fact that all the ionisable
residues are protonated might actually slightly destabilised the overall
structure of the protein. That is why I was willing to run the
equilibration with the protonation states allowed to change.
Another option could be to run the NPT equilibration with restraints on the
alpha carbons and no protonation changes allowed and then release the
restraints during the CpHMD NVT equilibrations on each replica (or first on
a single replica at pH 7 for example and do the split after this first
equilibration). But then I would expect some changes in density after
release of the restraints. Do you think the impact of the fluctuating
charges would be worse? I could also reduce the restraints to a minimum so
the release of them would not have to much impact I guess.
Thank you in advance,
Eric
On 12 November 2016 at 06:06, Jason Swails <jason.swails.gmail.com> wrote:
>
>
> On Nov 11, 2016, at 4:18 PM, Eric Lang <eric.lang.bristol.ac.uk> wrote:
>
> Dear Jason,
>
> Thanks a lot for your reply.
> Sorry, I did not make myself clear: I know that I cannot run REMD in NPT a
> that each replica needs to have exactly the same box size.
>
> What I would like to do after minimisation and heating is to run a
> *single* replica at pH 7 in NPT allowing the protonation states to change
> (so effectively NPT CpHMD). This would be the equilibration run. I would
> therefore end up with an equilibrated structure in which all the ionisable
> residues won't be necessarily be protonated. I then want to use this
> structure as a starting point for all the replicas for pH-REMD, which would
> be run in NVT (so I would do the split at this stage after NPT
> equilibration at constant pH).
>
> What I would like to know is:
> - Is it possible to run CpHMD (not pH-REMD) in NPT?
>
>
> Yes, but you run the risk that the fluctuating charge can throw off the
> density due to finite size effects (the pressure and free energy is
> sensitive to net charges when using the net neutralizing plasma like Amber
> does.
>
> If you can, I would suggest running the neutral system until it reaches
> equilibrium and then switching to CpHMD.
>
> - Is it a problem to have, for each replica of the pH-REMD, a starting
> structure that does not have all the protons (and is protonated according
> to a given pH) or is there a risk that this will lead to a poor protonation
> state sampling ?
>
>
> That should be fine. That was common practice for me.
>
> HTH,
> Jason
>
>
> Many thanks in advance,
>
> Eric
>
> On 11 November 2016 at 17:39, Jason Swails <jason.swails.gmail.com> wrote:
>
>>
>>
>> On Fri, Nov 11, 2016 at 12:31 PM, Eric Lang <eric.lang.bristol.ac.uk>
>> wrote:
>>
>>> Hello,
>>>
>>> I want to run some explicit solvent pH-REMD simulations, and I just want
>>> to
>>> check if, after the heating phase, I can run a NPT equilibration at a
>>> constant pH with the protonation states allowed to change.
>>>
>>
>> NpT + REMD in Amber is not supported for any kind of REMD. Furthermore,
>> every replica *should* have exactly the same box size when doing REMD with
>> PBC. This is because the exchange criteria does not incorporate
>> volume/pressure contributions, as I believe would be required to correctly
>> sample from the desired distribution.
>>
>> The proper procedure to use is to run a *single* replica until you get a
>> stabilized and sensible density, then build each replica from that starting
>> point using NVT the rest of the time.
>>
>> HTH,
>> Jason
>>
>> --
>> Jason M. Swails
>>
>
>
>
> --
>
> Eric Lang
>
> BrisSynBio Postdoctoral Research Associate Modelling
> Centre for Computational Chemistry
> School of Chemistry - University of Bristol
> Bristol BS8 1TS - United Kingdom
>
>
--
Eric Lang
BrisSynBio Postdoctoral Research Associate Modelling
Centre for Computational Chemistry
School of Chemistry - University of Bristol
Bristol BS8 1TS - United Kingdom
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Nov 14 2016 - 02:30:02 PST
