[gmx-users] :a grompp problem using a ffamber port
feenstra at few.vu.nl
Thu Oct 27 09:38:30 CEST 2005
David Mobley wrote:
> You might want to correspond with the Pande group on this, as they're the
> ones who put out the port.
> I had a similar problem with Cl- ions and, as a temporary solution, modified
> the ions.itp file in the top directory to have an #ifdef _FF_AMBER section
> specifying the appropriate ion definitions (copied from earlier in the
> ions.itp file but changing the atom types to the amber atom types -- i.e.
> CL- is amber94_30). My conclusion was that no ion parameters were being
> loaded when _FF_AMBER was defined, so I thought this was necessary. I
> remember corresponding with Eric Sorin (Pande group) on this and he had some
> alternative method of setting up systems where this wasn't a problem, I
> Anyway, you'll probably need to talk to them, as the Amber FF is not a part
> of Gromacs, so in general you probably won't get support from this list.
Aside from that, using the parameters from ions.itp, which are intended
to be used with the Gromos forcefields (that is why the #ifdef is
there!), with the Amber ff instead is probably a bad idea. I'd recommend
checking the parameters carefully with the ion parameters from Amber
itself (beware of unit conversions, e.g., A vs nm and cal vs J).
* NOTE: New Affiliation, Phone & Fax numbers (below) *
| | |
| _ _ ___,| K. Anton Feenstra |
| / \ / \'| | | IBIVU/Bioinformatics - Vrije Universiteit Amsterdam |
|( | )| | | De Boelelaan 1083 - 1081A HV Amsterdam - Netherlands |
| \_/ \_/ | | | Tel +31 20 59 87783 - Fax +31 20 59 87653 - Room P440 |
| | Feenstra at few.vu.nl - www.few.vu.nl/~feenstra/ |
| | "If You See Me Getting High, Knock Me Down" (RHCP) |
More information about the gromacs.org_gmx-users