swart_at_few.vu.nl wrote:

>I think that at the moment it is not yet possible, to do what you wanted to,
>unless you specify everything in Z-matrix format. Is there any specific reason
>why you don't want to do it in Z-matrix ?
>

The reason is, that I want to cut out a part of a protein, geometry
optimize it and afterwards put it back into the protein. Therefore I
need to fix the atoms which are linked to the rest of the protein in
cartesian space.
Maybe I can do it as a kind of QM/MM (with all atoms fixed in the MM
part and give the QM part in Z-matrix)?
Another reason why I am not very eager to do it in Z-matrix is that my
QM part has 112 atoms, 12 atoms need to be fixed in cartesian space. I
have not much experience with Z-matrix, but it seems to me fairly
complicated compared to dumping in the cartesian coordinates from PDB...
- Maybe I am lazy! :-)
I also have no experience with QM/MM and can't estimate how complicated
it is to prepare the input for a metallo-protein.

My first idea was also to use ZCart, as Paulo suggests, but then I read
"One should then use the ZCart format: give Cartesian coordinates and
supply the structure of the Z-matrix. Again, in this case you cannot use
constraints." (User Manual 2002.03, p. 42).

Serguei's suggestion sounds promising to me, but might be a bit painfull
for such a big system. Has anybody a script to convert PDB to Z-matrix,
which would do (part of) the job?

Thanks,

Timm
Received on 2003-02-26 18:49:16