We need to have, at least, entity ids and we need to follow the requirement that every atom in the atom_site loop has a unique id field. It does not, surprisingly, have to be sequential or numeric. It can be any unique string (this may be useful for very large proteins or large numbers of models, and therefore we may want to use alphanumeric strings)
The easiest solution to making sure all header information is present would be, for now, to copy it from the original input cif file.
We need to have, at least, entity ids and we need to follow the requirement that every atom in the atom_site loop has a unique
idfield. It does not, surprisingly, have to be sequential or numeric. It can be any unique string (this may be useful for very large proteins or large numbers of models, and therefore we may want to use alphanumeric strings)The easiest solution to making sure all header information is present would be, for now, to copy it from the original input cif file.