[Molecularmechanics] Request for comments

[Peter Murray-Rust]

At 19:07 07/11/2003 +0100, Konrad Hinsen wrote:
>On Friday 07 November 2003 18:29, Peter Murray-Rust wrote:
>
>
> >          Since I do not implement forcefield programs I'll take a neutral
> > position between the two proposals, but note that David's is closer to the
> > crystallographic topology files as it combines chemical constitution and
> > perception for forcefields.
>
>What do you mean by "crystallographic topology" files, and how are they
>related to forcefields?
They are used in constrained refinement and simulated annealing/MD methods 
for Xray refinements (tools like X-PLOR are direct descendants of (I think 
CHARMM)).

Crystallographers have to create these topological dictionaries for each 
new fragment - cf the HET dictionary of the PDB. They have to define the FF 
parameters (or pseudoFF parameters) for each HET group. The relevance is 
that if you have an unusual fragment in the molecule then it may well have 
been described in this way. It probably reflects that my experience has 
been slightly more in this field that unconstrained MD.

P.