[gmx-users] k-space settings in GROMACS

[Anh Vo]

Hi Mark,


Thank you very much for your quick response.

Would the command   'ewald-rtol   =  1e-05'  do the same job?


I'm not sure because the 1e-5 in  'kspace_style    pppm  1e-5'  defines how
much the RMS error is smaller than the reference force, while 1e-05 in
'ewald-rtol   = 1e-05'  defines the relative strength of the Ewald-shifted
direct potential.


Also, would there be much difference (e.g. in the energy of the molecules,
in the output pressure, etc.) when PME and P3M are used?


Thank you,

Anh Vo



--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------



Date: Wed, 18 Mar 2020 07:46:23 +0100

From: Mark Abraham <mark.j.abraham at gmail.com>

To: Discussion list for GROMACS users <gmx-users at gromacs.org>

Cc: Discussion list for GROMACS users

                <gromacs.org_gmx-users at maillist.sys.kth.se>

Subject: Re: [gmx-users] k-space settings in GROMACS

Message-ID:

                <
CAMNuMAStNtbe8s-QdkChTUBdM-ZrX_ZYdxpGVT-eT--nE2iWAA at mail.gmail.com>

Content-Type: text/plain; charset="UTF-8"



Hi,



On Wed, 18 Mar 2020 at 07:02, Anh Vo <atv55 at msstate.edu> wrote:

> Hi everyone,

>

> I'm trying to reconstruct in GROMACS some LAMMPS simulation to compare

> between the two. I'm simulating a lipid bilayer system with 72 POPC lipids

> and 9072 water molecules, as well as a single lipid system (1 POPC

> molecule) and water only system (100 H2O molecules).

>



Start with a simpler system when proving you understand how to set up

equivalent simulations.



I try to use the same settings in GROMACS as in LAMMPS. However, in LAMMPS

> I have *kspace_style* to define a long-range solver and *kspace_modify* to

> set

> parameters used by the kspace solvers; but I'm not sure which command I

> should use for similar settings in GROMACS. (I have read the manual and

> searched online but haven't figured it out. Please accept my apology if

> this seems to be an obvious question).

>

> In details, these 2 commands in LAMMPS are:

> *  - kspace_style    pppm  1e-5   *(pppm is P3M algorithm, 1.0e-5 means

> that the RMS error will be a factor of 100,000 smaller than the reference

> force)

>



Then you'll know that ewald_rtol does exactly that, and that

http://secure-web.cisco.com/1bK_dF9SMkoS8dyT5J_HpebY83O47XvhPiIf89xavsJV9dlNvLV7TtA-4DpVUJEkHQGTHa44uJzJBJ807pEGqJfRzXfnyD5srJY1X8Nhys5FdF_nE7ANfoc-7bT5mI2rXACvnYfCjPkft84Do34WkmYwMK1DmKKvmDwjhfiRsZYNEVopvDcmeJcwVnGwLUBbnKOLBPlsvDzkRU9-YHvOZ7pdvASO-6UOuELXEiuPOrFBjxgVL1oY8eOxlKM7C6DE4eUhTZ6K5m-ygWc7iLjDlxbqIkEMSkQx1EFixWuxWnlMum4BQ05zUNCyLGHHudIx7/http%3A%2F%2Fmanual.gromacs.org%2Fdocumentation%2Fcurrent%2Fuser-guide%2Fmdp-options.html#mdp-value-coulombtype=P3M-AD

does

what you want.



*  - kspace_modify   diff  ad    *(*diff*  keyword specifies the

> differentiation scheme used by the PPPM method to compute forces on

> particles,  *ad*  keyword means approach uses only 1 FFT to transfer

> information back to real space for a total of 2 FFTs per timestep).

>

> Please help me to get similar settings in GROMACS (e.g. which command

> should I use ?). Also, would it make a significant difference (e.g. in the

> energy of the molecules, in the output pressure, etc.) if I use PME
instead

> of PPPM in GROMACS ?

>



No need to do that.



Mark





> Thank you very much for your help.

>

> Best,

> Anh Vo