On 1/01/2012 7:45 AM, Lara Bunte wrote:
Hello
I have two questions.
First question:
If you parameterize a force field (which means that you make changes
in atomtypes.atp, create a new .rtp and a .hdb file, make changes in
ffnonbonded.itp, create a residuetypes.dat file and finally make
symbolic links to gromacs) and you use pdb2gmx to get a topology file
(topol.top) and this file has not all informations or something is
wrong what is better:
1.) Try to fix missing or false information in the topology file
(detective work in the topology)
or should one
2.) Try to fix something in the created and modified files of the
force field and parametrization
If you have done a complete parameterization and constructed a complete
way for pdb2gmx to built a topology from a coordinate file then the
problem cannot arise. If you haven't done the former, then go back and
do it. If you haven't done the latter, then it is probably right to fix
the contents of the force field directory (rather than the produced
topology) on the theory that you may want to use the force field again
some time.
Second question:
This question is implied in someway: If everything in the
parametrization is correct will be the topology you get by pdb2gmx
alway correct for further calculations?
To summarize: If someone has bad (even false) topology, should you try
to fix the topology or is a bad topology always a strong
identification for bad parametrization?
They're orthogonal. The topology expresses a particular combination of
the parameters that implements a model. Both topology and parameters
have to be right for the model to work, and failure doesn't localize the
problem to one or the other aspect. Just as with writing, even if the
parser (grompp) thinks the words you are using make grammatical sense
(no error, so sufficient parameter values and connectivity), only you
can judge whether the way in which you have followed the rules should
make for a meaningful sentence (valid model).
Mark
--
gmx-users mailing list gmx-users@gromacs.org
http://lists.gromacs.org/mailman/listinfo/gmx-users
Please search the archive at
http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
Please don't post (un)subscribe requests to the list. Use the
www interface or send it to gmx-users-requ...@gromacs.org.
Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
