Re: NMR entries

• Messages sorted by: [ date ][ subject ][ author ]
• Previous message: Eldon Ulrich: "NMR entries"

This note is in response to the message:
It has been suggested that the data tag '_atom_site.label_alt_id' could be used
to solve the problem described above without additional data tags being
created. Is there a specific problem or objection with implementing the
'_atom_site.label_alt_id' suggestion?  One objection has involved calculating
atom site occupancies.  However, the '_atom_site.occupancy' tag is not
mandatory and does not need to be included for NMR structures.

Eldon Ulrich

---------------------------------------------------------------------------

The most important thing would be to make a decision of some kind.
Since there are some technical problems with using
_atom_site.label_alt_id without a redefinition of the token and
some clarifying language on what is happening with
_atom_site.occupancy in the case of NMR entries, and one can get
into rather complex questions on the handling of disorder, vs
microheterogeneity vs. multiple models, temporarily creating a
token to simply carry the existing information in mmCIF versions
of existing PDB entries, without redefining anything now in the
dictionary would seem a conservative holding action while waiting
for a full technical resolution of the proper handling of NMR entries.

Please note that a implementation of this approach does not even
require an official change in the mmCIF dicitionary, since the
extra token could be treated as a user-defined token listed in
the PUBL_MANUSCRIPT_INCL category, but a generally agreed
approach would be better for all concerned.

Certainly, if the work has already been done on augmenting the
definition of _atom_site.label_alt_id with examples of how
NMR entries are to be handled, I, for one, would be happy to
switch over to using that, but I don't recall seeing that
in the latest dictionary nor in the discussions on the subject.
My apologies if I missed it.
  -- H. J. Bernstein

• Previous message: Eldon Ulrich: "NMR entries"