[gmx-users] Trying to use AMBER parameters in GROMACS

[Qinghua Liao]

Hello,

You could try to build the system with AmberTools (it is free), choose 
the force fields available in Amber.
Once you get the topology and coordinates files (prmtop and inpcrd), you 
can convert them into top and gro
in Gromacs format with *parmed*. Then you can perform the simulations 
with Gromacs.

import parmed as pmd

#convert prmtop and inpcrd into top and gro
amber = pmd.load_file('mol_sol.prmtop','mol_sol.inpcrd')
#
amber.save('mol_sol.top')
amber.save('mol_sol.gro')

You can google for more details.


All the best,
Qinghua

On 01/03/2018 05:52 PM, Crystal Vander Zanden wrote:
> Dear GROMACS Community,
>
> I am trying to parameterize my system using AMBER force fields, and then do
> molecular dynamics with the GROMACS software.  I want to use the Lipid17
> force field, which is not build into GROMACS.  Does anyone know of a way to
> do this?
>
>
> I have found 3 programs people have written to do this, but in short, none
> of them work because the codes are too old (ca. 2003).
>
> 1) https://code.google.com/archive/p/amber2gmx/downloads
> This would work, but it is written to use a command line parser package
> that is too old and I can’t find (com.tsd.apps).
>
> 2) The amber2gmxrtp script build into GROMACS
> This doesn’t work because it was written to take .in and .prep files (made
> for the antiquated AMBER Prep program).  Those file types have been
> replaced with .lib (made for Leap)
>
> 3)The amberconv.tgz package from the GROMACS website (
> http://www.gromacs.org/Downloads/User_contributions/Other_software)
> This was missing an include file when I tried to compile it.
>
>
>
> The reason for using AMBER force fields:  My system has lipids, protein,
> and a small molecule - all of which can be well described with AMBER force
> fields.  Since AMBER force fields are QM-based, it would be really easy to
> parameterize the small molecule using Antechamber and low-level QM.
> GROMACS force fields are empirically derived, so it will be much more
> challenging to parameterize the small molecule in a way that’s consistent
> with the rest of the force field for lipids and protein.  I plan on
> simulating more small molecules in the future.
>
> I am hoping to use GROMACS because it's free, fast, and also not too
> difficult to learn.
>
>
> Thanks in advance for any advice you’re able to provide!!
> Crystal
>