Menu
▾
▴
#219 write mmCIF
As far as I know it is currently not possible to write the model data from mmCIF format files exceeding the PDB format limits with Jmol while preserving the PDB/mmCIF-specific properties.
There is a growing number of such entries in the PDB (currently presumably about 400). Since the mmCIF format is the recommended (and only) format for reading those models it would be consistent to be able two write them also in mmCIF format, e.g. after modifying the data.
It would also be helpful for large non-PDB structures.
So the basic question is: Of all the information in a PDB or mmCIF file, what is the minimum subset that would be appropriate. It's easy enough to spit out atoms and connctions and probably secondary structure. Biomolecule symmetry might be a different story. So what is critical do you think?
I think it should at least be the same information that is written currently with 'write PDB ...'.
As far as I know this includes the ATOM,HETATM and CONECT information and (hopefully) the MODEL information.
I would expect that most of the changes applied using Jmol will be rather in these records than somewhere else. Since it is quite easy to add the original "header" data before writing the data into a file, also for mmCIF, these should be sufficient for most applications.