[IUCr Home Page]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Final draft of Phase-ID report

Dear David

Thank you for the final draft of the phase identifiers report. I approve
the contents of the report, subject to some possible editorial amendments
in light of my comments below. I can see no reason why the IUCr should not 
make use of the proposed identifier in its journals and other

(1) I understand that the IChI is now known as the "INChI" (IUPAC-NIST
Chemical Identifier) in recognition of the important contributions by
members of NIST. That's not clear from the IUPAC Project page and I
haven't found an authoritative site describing its recent evolution, but
the article by Steve Heller (http://www.hellers.com/steve/resume/p159.html)
talks a little bit about this.

(2) The example and description in Section 5.1 uses 1.00Beta as the
I(N)ChI version identifier, but the examples of Section 8.2 use version
1.02. These should be reconciled.

(3) Two remarks about your gloss on the example of Section 5.1.
  (i) I found the sentence "(H1 is deuterium)" confusing. Should this be
      "The atom at site 1 is deuterium" ? And while D is readily
      understood as the conventional symbol for hydrogen-2, how does one
      indicate carbon-14 etc?
  (ii) I am guessing that the second occurrence of the SC: field is 
      associated in sequence with the preceding I: field (i.e. it gives
      the stereochemistry of the isotopic form with deuterium). This
      indicates an order dependence in the components of the INChI
      given in this form. While one doesn't want to have to describe the
      INChI in too much detail, is this point worth discussing?

  Alternatively, might the I: and second SC: field be dropped from this

(4) Possibly the terse explanations of the INChI tags should be expanded a 
little bit, if only to expand the abbreviated words.

(5) Should Section 6.2 end with the sentence "Quasicrystals are not
considered further by this Working Group."? Or have we anything to say
that might guide further extensions to include quasicrystals?

(6) In Section 7.2.3 is stated

"Only the lower space group number of each pair should be used. 
 The chirality is often not determined and is only significant if
 the crystal contains a molecule whose chirality is described
 elsewhere in IChI."

I find this a little confusing. If one is describing a phase involving a
molecule of *known* and *fully described* chirality in a chiral space
group of higher number within the pair, then surely that higher
space-group number is the one that should go into the identifier?
You are, after all, identifying a very specific and well-characterised
phase in this case.

Is it not better to recommend that where the molecular chirality is not
described or known, *then* the lower of the two possible space groups
should be assigned? Likewise, any search query that includes a
space-group number should for the chiral space groups also generate a
query with the other space-group number to search for all possible chiral
assignments (including mistaken ones).

In any case, I think it would be helpful in the submitted report to
collect these space-group pairs together in a table so that they stand out 
more clearly and are easy to locate.

(7) Lastly, the use of the identifier in this form will depend on the
cooperation of the INChI community. Although we have good reason to
believe that this will be forthcoming, it might be helpful also to
formalise the components of the phase identifier as core CIF items
in their own right, so that they can be used independently of the INChI
connection. One imagines a PHASE_IDENTIFY category with items such as
These need not be specified in this report, but the Working Group might
wish to refer this question to the core CIF working group and mention in
the report that this is under consideration.

Please accept my apologies for not attending more closely to the earlier
rounds of discussions on this topic. In reviewing what has gone before, I
have found much fruitful discussion and the seeds of several ideas for
extending this formalism to accommodate ambiguous cases in the
future. Nevertheless, I am impressed by the compactness of the final
proposal and its potential for integration with INChI and the CCN phase
transition nomenclature proposals.

Best wishes

PS: (With apologies to Sydney Abrahams for my tardiness in mentioning
this.) If other members of the Working Group wish to review past
discussions on the IUCr web site, they will find that the original URL
no longer works, because the machine hosting that discussion has been 
disconnected from the Internet. The list archive can instead be found at
phase-identifiers mailing list

Reply to: [list | sender only]

Copyright © International Union of Crystallography

IUCr Webmaster