From: Francesco Pietra (
Date: Wed Jun 13 2018 - 09:30:15 CDT

Hi Josh: referring to the way FEPs can be run with NAMD, it is sufficient
to have a system running MD, to which flags for FEP can be added. I hope
the same applies to GROMACS. If so, no need to extend TOPOGROMACS. I have
first to become familiar with GROMACS, where plenty of studies of
protein-ligand FEP have appeared in the literature.

Thanks again

On Wed, Jun 13, 2018 at 4:10 PM, Vermaas, Joshua <>

> Hi Francesco,
> Coordinates are easy. They are passed along with the -c option to grompp.
> Velocities... Probably its possible somehow once you convert the file
> types. The difficulty is going to be that the tpr file that grompp
> generates normally doesn't (and maybe cannot?) include velocities. Instead,
> input velocities are normally passed directly to mdrun, so that is
> something outside of the TopoGromacs pipeline. FEP specification is also
> not covered by TopoGromacs. In GROMACS, FEP is specified at the level of
> the topology file. I didn't want to deal with it when I was writing
> TopoGromacs, and my own use cases were all equilibrium simulations anyway,
> so you may need to edit the code to correctly handle FEP.
> -Josh
> On 2018-06-13 07:42:30-06:00 Francesco Pietra wrote:
> Hi Joshua: Following your indications and a preliminary reading of your
> paper, I applied topogromacs to my charmm36-parameterized system of a
> protein with GDP ligand and a diterpenoid ligand (the latter defined in a
> .str streaming file, without the need of a separate .a prm file) getting
> the .top file with apparently all components. A most astonishing conversion!
> I am still unfamiliar with GROMACS (planning to use it to compare FEP
> simulations with NAMD) but may I boldly ask whether passing .the .top file
> to grompp there is a way to preserve the velocities, or at least the
> equilibrated coordinates, of NAMD?
> Thanks a lot, also to John
> francesco
> On Wed, Jun 13, 2018 at 11:53 AM, Vermaas, Joshua <
> > wrote:
>> Yup, as John described, normally a user does NOT need to change anything
>> in this folder if they are running a 1.9.4 alpha or 1.9.2. The regular
>> workflow works where you load your namd psf file, and then execute the
>> commands described in the paper (10.1021/acs.jcim.6b00103):
>> package require topotools
>> topo writegmxtop <outputname>.top [list prmfile1.prm prmfile2.prm ...]
>> When running on 1.9.3, you may need to fix a syntax error that was
>> introduced right before release in the acknowledgements in the directory
>> John specified (see
>> ist/vmd-l/28199.html).
>> <>
>> -Josh
>> On 2018-06-11 15:36:21-06:00 wrote:
>> Hi,
>> If you look in the VMD installation folder, you'll see that the topogromacs
>> code is found within this subdirectory:
>> plugins/noarch/tcl/topotools1.7/topogromacs.tcl
>> Now, that being said, you should just run "package require topotools" to get
>> access to all of the associated commands.
>> Best regards,
>> John Stone
>> On Mon, Jun 11, 2018 at 10:30:51PM +0200, Francesco Pietra wrote:
>> &amp;amp;gt; I would like to try porting to gromacs/charmm36 a protein-ligand system
>> &amp;amp;gt; built and running on namd/charmm36. To this end I was unable to find a
>> &amp;amp;gt; topotool folder, inclusive of topogromacs, on my installations of vmd
>> &amp;amp;gt; 1.9.4a12 or vmd 1.9.3. Although I know how to add third-part plugins to
>> &amp;amp;gt; vmd, I am confused here. Thanks for advice.
>> &amp;amp;gt;
>> &amp;amp;gt; francesco pietra
>> --
>> NIH Center for Macromolecular Modeling and Bioinformatics
>> Beckman Institute for Advanced Science and Technology
>> University of Illinois, 405 N. Mathews Ave, Urbana, IL 61801 <>;amp;amp;;amp;amp;sdata=Si4CHOUJ%2FGEedPBReaZmLFohgkTSuXYLE8c%2Bd5FL1%2BQ%3D&amp;amp;amp;reserved=0 Phone: 217-244-3349;amp;amp;;amp;amp;sdata=eFyk21SZAIlk%2Ba%2FS38My78ul2f8LyZeYZYDzoaMqrNY%3D&amp;amp;amp;reserved=0