Data items in the PDBX_CONSTRUCT category specify a sequence of
  nucleic acids or amino acids. It is a catch-all that may be used to
  provide details of sequences known to be relevant to the project as well
  as primers, plasmids, proteins and such like that are either used or
  produced during the protein production process. Molecules described
  here are not necessarily complete, so for instance it would be
  possible to include either a complete plasmid or just its insert.
  This category may be considered as an abbreviated form of _entity where
  the molecules described are not required to appear in the final co-ordinates.

  Note that the details provided here all pertain to a single entry as defined
  at deposition. It is anticipated that _pdbx_construct.id would also be
  composed of a sequence that is unique within a given site prefixed by a code
  that identifies that site and would, therefore, be GLOBALLY unique. Thus
  this category could also be used locally to store details about the different
  constructs used during protein production without reference to the entry_id
 (which only becomes a  meaningful concept during deposition).