*** For details on how to be removed from this list visit the ***
*** CCP4 home page http://www.ccp4.ac.uk ***
Dear All,
We have recently "solved" a crystal structure using Se-Met phasing and
are refining the model against native data at 2.25 A. The symmetry is
primitive orthorhombic and the data seem to have all the systematic
absences expected for P212121 (cell: 43.550 121.417 143.785 90.00
90.00 90.00). Phil Evans' program POINTLESS also thinks it is P212121.
The monomer size is about 18 kDa and we have two dimers per ASU (sol
content ~55%). The current model contains 83% of the sequence in each
chain, and some 400 waters have been added. Although there is still a
little room for improvement in the model, the FreeR appears to be stuck
at ~34%. The two dimers within the ASU are in essentially the same
orientation and related by a translation that is approximately, but not
exactly, 0, 0.5, 0 (see also SFCHECK output below) - this shows up as a
strong peak in a native Patterson. Also, the NCS 2-folds are parallel
to the z-axis. It sounds like this may be similar to cases that were
described under the "Rfree can not be decreased" posting a few weeks ago
(especially the example described in Acta Crystallographica Section D:
Biological Crystallography D60, 447-452). We have tried taking the
current model and doing MR in different space groups: monoclinic (both
P2 and P21 with all possible 2-folds/2-fold screws) with 8 subunits per
ASU is no better. Running PHASER in all possible primitive orthorhombic
space groups finds a solution in P21221, that has essentially the same
crystal packing, but this gave higher Rfree values than the original
P212121. Curiously, re-solving in P212121 by MR gives a solution that is
shifted with respect to the original by around 5Ang. From an inspection
of pseudo precession plots there appears to be some pseudo-centring on
the 0kl plot, at least at low resolution, the spots being systematically
weaker where k+l is odd, and the odd l lines are generally weaker.
SFCHECK suggests there is some anisotropy and also gives a rather high
perfect twinning test score of 2.65 (TRUNCATE gives an acentric 2nd
moment of 2.52).
SCALA summary . . . .
Overall OuterShell
Low resolution limit 37.24 2.37
High resolution limit 2.25 2.25
Rmerge 0.041 0.159
Rmeas (within I+/I-) 0.050 0.214
Rmeas (all I+ & I-) 0.050 0.214
Rpim (within I+/I-) 0.028 0.142
Rpim (all I+ & I-) 0.028 0.142
Fractional partial bias -0.004 0.017
Total number of observations 103192 7284
Total number unique 34792 3660
Mean((I)/sd(I)) 18.3 5.4
Completeness 94.0 70.8
Multiplicity 3.0 2.0
OK, the completeness and multiplicity are not as high as we would have
liked, but we ran out of beamtime!
SFCHECK output . . . .
--- Structure factors ---
Resolution /input data/ : 32.23 - 2.25
Number of reflections : 34645 ( in_file : 34694)
Number of unacceptable reflections : 49
Number of unacceptable reflections as systematycal absent : 49
Number of reflections with I > sigma(I) : 34417
Number of reflections with I > 3sigma(I): 29652
R_stand(I) = <sig(I)>/<I> : 0.040
Number of acceptable reflections: 34645
for resolution : 32.23 - 2.25
Optical Resolution: 1.74
Expected Optical Resolution for complete data set: 1.74
/ Optical resolution - expected minimal distance between
two resolved peaks in the electron density map./
Resmax_used(opt): 2.78
Number of max possible refls : 37204 actual : 34645
Completeness : 93.1 %
R_stand(F) = <sig(F)>/<F> :0.032
Boverall /estimated by Patterson/ : 33.98
Minimal estimated error : 0.0120
Pseudo-translation is detected (%): 21.9
pseudo-trans.vector /frac/: 0.000 0.500 0.014
Perfect twinning test <I^2>/<I>^2 : 2.6505
Anisotropic distribution of Structure Factors:
Ratio of Eigen values : 1.0000 0.9268 0.7481
Should we expect high Rfree values with this kind of pseudo-translation
and would this account for the high perfect twinning test score?
Is it possible that the crystal is actually a mixture of P212121 and
P21221 domains?
Although I suspect the only solution is to produce another crystal form,
any suggestions as to how to proceed with these data would be greatly
received.
Many thanks
Dave Lawson
-------------------------------
Dr. David M. Lawson
Biological Chemistry Dept.,
John Innes Centre,
Norwich,
NR4 7UH, UK.
Tel: +44-(0)1603-450725
Fax: +44-(0)1603-450018
Email: [EMAIL PROTECTED]
Web: http://www.jic.bbsrc.ac.uk/staff/david-lawson/index.htm
