[gmx-users] Pressure control with constraints incompatible with expanded ensemble

[Michael Shirts]

Hi, Jacob-
Unfortunately, that is correct.  We had though we'd have functionality that
could support this combination in 2018 (a Monte Carlo barostat and/or
expanded ensemble in alchemical space working with leapfrog), but those did
not make it in.  The support for MTTK with constraints was removed in order
to simplify the inner loop of MD for other improvements.

The good thing is that in virtually all situations at P near 1 atm, \Delta
A and \Delta G for solvation calculations are pretty much identical, since
P \Delta V is so smal.  We can chat more if you think that you would need
additional help on this.  We are working with some improvements for
expanded ensemble, so contact me off the list if you want to discuss those
and see if they will help.

We DO know for sure that expanded ensemble or replica exchange with
Berendsen pressure control has some nasty issues (will be mentioned in a
paper hopefully being submitted soon, so it's better documented). So don't
do that.



On Wed, Sep 5, 2018 at 12:54 PM Jacob Monroe <jimonroe at umail.ucsb.edu>
wrote:

> Hi all,
>
> I would like to use Gromacs (2018 versions on GPUs) to compute solvation
> free energies using expanded ensemble simulations.  I have no problems in
> the NVT ensemble, but when I try and implement pressure control I run into
> errors.
>
> Specifically, if I want to use expanded ensemble, I must use the md-vv
> integrator.  If I use this integrator, I must select MTTK pressure control
> (I’d rather not use Berendsen for production runs), but if I want to
> constrain hydrogen bonds, I cannot use MTTK (grompp throws an error).
> Previous versions of Gromacs appear to have allowed MTTK as long as SHAKE
> was used for constraints (see the mdp files here:
> http://www.alchemistry.org/wiki/GROMACS_4.6_example:_Ethanol_solvation_with_expanded_ensemble
> ).
>
> So is the only way to run NPT, expanded ensemble simulations with current
> versions of Gromacs to use the Berendsen barostat?
>
> Thanks,
> Jacob
> --
> Gromacs Users mailing list
>
> * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
> posting!
>
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>
> * For (un)subscribe requests visit
> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
> send a mail to gmx-users-request at gromacs.org.