Am 23 Jul 2009 um 0:31 hat George M. Sheldrick geschrieben:
> Dear Matthew,
>
> If it is P43 and merohedrally twinned, you should not see systematic
> absences for the 21 axis. In fact there is no way of getting the
> P43212 systematic absences from a twinned crystal of lower
> symmetry.
> I presume that you also tried MR in P41212.
>
> Best wishes, George
>
Dear George,
is this really true?
We had a case of P43212 (also indicated by the systematic absences on the P21 axis) in
which in the final refinement rounds some regions in the structures remained "disordered"
(PDB 3FWQ). These were functionally important regions; so we made some efforts to
improve them. Among others we reduced the symmetry to P43 and treated the crystal as a
perfect twin. Of course, we had then a larger asymmetric unit and two chain in the P43-AU
rather than one chain in the P43212-AU.
Afterwards the density in the critical regions was much better. This was true for both chains,
however, with some structural differences between them. So my impression was that in this
case the reduction of the crystal symmetry combined with twinning was a good strategy to
come to a better defined electron density and structure. Or was it a case of "over-twinning"?
Many thanks in advance and best wishes from
Karsten
-------------------------
Dr. Karsten Niefind
University of Cologne
Department of Chemistry - Institute of Biochemistry
Zuelpicher Str. 47, D-50674 Cologne
Tel.: ++49 221 470 6444
Fax: ++49 221 470 6431
|