Twinning just requires some geometric relation between crystal axes to
yes - it is possible in p1
othercell will tell you if there is some other indexing which will
generate a similar arrangement of axes.
It could be possible that it has higher symmetry of course.
Eleanor
José Trincão wrote:
> Dear all,
> we have a protein isolated from mouse liver which crystallized in P1. The amount of protein was very little so we could not get better crystals. The protein expressed in E.coli did not yield any usable crystals.
>
> We managed to collect data from 2 crystals after annealing at the beamline (before annealing the data was useless). The data is very anisotropic, useless for about 50º (split crystal), but looks ok on other orientations.
> The data collected could only be processed in P1 with XDS - mosflm consistently crashes in autoindexing.
> The protein is ~150kDa - we could find 4 mol in the unit cell by MR.
> Although the data is ~95% complete to ~3 A, it extends to >2.5 (~60% completeness, and the maps are significantly better if we include these).
> All the tests indicate twinning. Even then we could refine the structure to R ~25% and Rfree ~33% but are stuck at this point. Only 55 waters could be found (for 5400 residues).
> My questions are:
> - is it possible to have twinning in P1? Can it be detwinned?
> - if we can't get better crystals, will it ever be possible to convince a referee that the high Rfree is due to twinning and that the model is ok?
>
> - apart from getting better crystals, is there anything we could try to improve our data, lower our Rfree?
>
> Thanks!
>
> Jose Trincao
> José Trincão, PhD CQFB@FCT-UNL
> 2829-516 Caparica, Portugal
>
> "It's very hard to make predictions... especially about the future" - Niels Bohr
|