Perhaps someone should contact the PDB first and get their opinion on
this. They may not be aware that their practice diverges from IUCr and
Cambridge data base. Centrosymmetric space groups are something of a
novelty in the macromolecular world. A quick search at RCSB turned up 13
entries.
On 01/18/12 12:04, Martyn Winn wrote:
On Wed, 2012-01-18 at 16:54 +0000, Paul Emsley wrote:
On 18/01/12 10:47, Martyn Winn wrote:
This needs a 1-line fix in syminfo.lib file. In the spacegroup entry
(the second entry in the file), replace:
symbol old 'P -1'
with
symbol old 'P 1-'
This works for me at least. The syminfo.lib file can be found in:
.../coot-whatever/share/coot/syminfo.lib
or in:
.../ccp4-whatever/lib/data/syminfo.lib
Yes, it is an mmdb issue, in that mmdb tries to match the spacegroup in
the CRYST1 card against the spacegroups in syminfo.lib (it checks
against "symbol xHM" and "symbol old" entries).
There are apparently a handful of centrosymmetric entries in the PDB,
e.g. racemic mixtures of synthetic peptides. Actually, looking at the
spacegroups offered in the RCSB advanced search menu, "P 1-" is the only
case.
Clearly, it is "only" a matter of convention whether you use "P -1" or
"P 1-" ....
Thanks for looking into this Martyn.
So just to be completely clear, CCP4 is changing symbol old for P -1 in
symop.lib?
Paul.
I have changed it in syminfo.lib not symop.lib
That is what mmdb uses nowadays, and seems to be what you distribute.
I have CVS'd the change into the CCP4 repository, and I presume it will
appear in the next release.
Sue Roberts pointed out to me that CCDC write out "P -1". syminfo.lib
now has:
symbol xHM 'P -1'
symbol old 'P 1-'
which should cover both cases (mmdb checks against both xHM and old).
m
--
=======================================================================
All Things Serve the Beam
=======================================================================
David J. Schuller
modern man in a post-modern world
MacCHESS, Cornell University
[email protected]
