[gmx-users] [ gmx-users ] RE: Vacuum Simulations

Martina Bertsch, PhD mbe404 at lulu.it.northwestern.edu
Wed Feb 11 19:07:00 CET 2004

Hi, Anton.

Thank you for your reply. I looked around for the conversion script you 
mentioned, but I only found the following package among the Contributions:
This is a program to convert GROMOS96 formatted topology, and 
conformation files to their GROMACS equivalents.
Uploaded 19:41 January 10, 2003 by Claudio M. Soares

I don't think that this is the script you were referring to, because I 
installed it and it seems to only be able to convert Gromos 96 
topologies to Gromacs.

I also found renumtop contributed by you, but there is no documentation 
except one gmx-user post. I tried it on an itp file created by PRODRG, 
but it returned the same atom types. Either that is not the script you 
had in mind or I am doing something wrong.

Could you give me the name of your script, or perhaps send it to me 
along with some information on how to use it to convert an ffgmx itp 
file to the ffG43a1 topology?



Anton Feenstra wrote:
Martina Bertsch, PhD wrote:

> Greetings.
> I would like to refine Autodock3 results of docking a nonpeptidic 
> indolic ligand to a membrane receptor model. Because it is 
> computationally expensive to simulate receptors in the membrane (the 
> 5-ns position restrained MD to relax the POPC takes me ~ 20 days), I 
> would like to use MD or simulated annealing in vacuum for some 
> intermediate refinement steps.
> I can generate the protein gro file suitable for vacuum simulations by 
> selecting the ffG43b1 parameter set in pdb2gmx. But, how do I generate 
> the ligand gro and itp consistent with this forcefield? I tried using 
> the ffgmx version of the ligand (as created in PRODRG) and editing the 
> vacuum ff to include the atomtypes, but somehow that does not seem 
> like the right thing to do.

No, you cannot mix atom types from different forcefields. So, instead
you must choose appropriate 'ff43a1' atom types for your ProDrg generated
'ffgmx' topology. Also, you will need to select appropriate 'ff43a1' types
for all bonds, angles, dihedrals and impropers (that is the way it must be
done in the Gromos forcefield). IIRC, I posted a script to the list (or
the contributions page), that will replace atom numbers in your .itp/.top
file by labels that include the atom type, as well as another script that
does the opposite. Having atom type info in stead of atom numbers makes
choosing bond-, angle- and dihedral-types so much easier. If you cannot
find these scripts, please ask me.

> Which energy groups do I specify for monitoring to estimate the 
> receptor-ligand binding energy: protein and drg or protein_drg?

protein and drg
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20040211/a14e1707/attachment.html>

More information about the gromacs.org_gmx-users mailing list