[gmx-users] All atom force field
k.a.feenstra at chem.rug.nl
Thu Oct 11 09:29:15 CEST 2001
David van der Spoel wrote:
> On Wed, 10 Oct 2001, Andrew Horsfield wrote:
> >Gromacs has the ffgmx2 all atom force field. This has a reduced range of
> >residues in its .rtp file compared with ffgmx. If I were to copy over the
> >residues from ffgmx to ffgmx2 I would have to replace compound atoms (CH3
> >etc) with separate ones. This creates the problem of assigning charges.
> >How important is it to get the charge assignments to individual atoms in a
> >residue correct (which I have to work out), as opposed to the net charge
> >on the residue (which I can get from the ffgmx file)?
> The ffgmx2 force field is not different from the ffgmx force field. The
> hydrogen atoms carry charge nor lennard JOnes parameters. In fact the
> whole force field was made only to be able to use distance restraints for
> NMR refinement.
In which case it would also make sense to create these hydrogens as
dummy atoms (use option -dummy hydrogens in pdb2gmx, see also J. Comput.
Chem. 20, 786 (1999)). This removes mass and motional freedom from hydrogen
atoms, in stead creating its position from that of some nearby heavy atoms.
Also, if the hydrogens are only necessary in analysis, it makes even
more sense to create them afterwards, using the program protonate.
protonate can process individual structures as well as whole trajectories.
It does, however, require the information present in the ffgmx2 forcefield,
so any residues in your system should also be present in ffgmx2, or should
be added as discussed above...
| | |
| _ _ ___,| Anton Feenstra Dept. of Biophysical Chemistry|
| / \ / \'| | | anton at chem.rug.nl University of Groningen|
|( | )| | | +31 50 363 4327 Nijenborgh 4 9747 AG Groningen|
| \_/ \_/ | | | http://md.chem.rug.nl/~anton The Netherlands|
| | "No One Could Foresee the End That Came So Fast" (Sl|
More information about the gromacs.org_gmx-users