write Chemical Component to cif file

[rimmartin]

Is it possible to populate a ChemComp or other structure in memory and then write to Chemical Component Dictionary cif?

Produce such standards as refmac monlib or Chemical Component Dictionary entries?

[wojdyr]

There is undocumented function add_chemcomp_to_block() for ChemComp->cif. I found a fragment of code that writes ChemComp to a file:

cif::Block new_block("out");
gemmi::add_chemcomp_to_block(cc, new_block);
gemmi::Ofstream os("ags.cif");
cif::write_cif_block_to_stream(os.ref(), new_block, cif::Style::Simple);

The same is possible in Python. The resulting cif file is more like Refmac dictionary than CCD.

As for populating ChemComp, it's possible from C++, but Python bindings are likely missing some of the functions. If you write more about what you want to do, we might be able to help you more.

[rimmartin]

Hi @wojdyr , that is promising:-) I'm at https://www.quantumbioinc.com/ where I write c++ code for better refinement and re-refinement of crystal structures. At the end of calculations we(or people using our software) want to output standards that can be used in the place of previous standards from refmac/ccd or donate them back to https://www.wwpdb.org/data/ccd Likely other people could use gemmi for similar or from raw crystal refinement.

I see https://project-gemmi.github.io/cxx-api/structgemmi_1_1Topo.html sets up restraints from reading standards from comp cif's. There would need to be algorithms implemented in gemmi for determining all the restraints from xyz structures/models? Or should apps do this on their own and populate ChemComp for io?

[wojdyr]

Normally, restraints are read from a monomer library such as the ones in CCP4 and in Phenix.

Gemmi has a function for making ad-hoc restraints based on the current geometry of a residue:

ChemComp cc = make_chemcomp_with_restraints(res);

but it's only to be used when the proper restraints are not available (because the user doesn't care enough to generate them). These ad-hoc restraints only stabilize the existing molecule's geometry. Does it answer your question?

[rimmartin]

Yes, I'll implement calling it that far and see if the people like the restraints it generates and the cif. Thank you