complicated.
Leaving aside the screwyness presumably it processed as PG422 OK.
Checks 1) NCS translation? 2) Twinned? ( both can be checked with
SFCHECK - it will tell you if there is pseudo translation and make a
guess at whether there is twinning.. The log file is terse but should
give you this info.)
Then check axial reflections - scala has given an indication, or use:
mtzutils hklin scala-op.mtz
AXIS h00 00l
This will list the axial reflections and might help you decide P41 21 2
or P43 21 2 or P4i 2 2
Are you likely to have one or two molecule in PG 422? It is possible
there is pseudo symmetry mixing up the 21 or 2 axes but that isnt
consistent with having a self rotation function which shoews anything
other than the crystal symmetry..
Eleanor
Schneider Sabine wrote:
Hi everyone,
I have got a dataset to 2.8A which according to Mosflm/XDS have P4
(etc) or P222 (etc) symmetry with a=b=71 c = 228 (same penalty in both
programs).
According to Matthews coefficient and self-rotation function (GLRF and
Molrep at kappa=180) I have 2 molecules in ASU, related through NCS.
Looking at systematic absences in hklview it seem to be P4. I
initially processed the data in P4 and pointless suggested P41 21
2/P43 21 2, so I reprocessed the data in P41 21 2.
As a molecular replacement model I used an ensemble of 3 structures
each with (unfortunately) only ~20% sequence ID and search was carried
out with one molecule. Residues that in the search model that were not
identical were stripped back to C gamma by chainsaw and the B-factors
were reset to 10 (mc) and 20 (sc).
I run Phaser searching in all alternative space groups and 2 molecules
in the ASU and it found a solution in P43 21 2. When I input the data
again in P43 21 2 into Phaser searching all spacegroups I get a
solution in P43 2 2? The difference in the two solutions is
association of the two molecules in the asymmetric unit along b. I
reprocessed the data in P43 2 2 and run phaser again against all space
groups and am finding P43 21 2?
I thought I might have mistaken the NCS as crystallographic symmetry
and have processed them in a to high symmetry space group. The data
are merging OK in P222 (Rmerge 5.3%) and P4322/P43212 (6%/6.5%).
I run Phaser with the data processed in P 2 2 2 against all possible
space groups and it found a solution in P2 2 21.
In all solutions (orthorhombic and tetragonal) I have clashes with the
symmetry molecules looking at the results in Coot.
The rotation function Z-scores and translation function Z-scores are
low as expected with 2.7 -2.9 and 4.7-5, respectively. And of course
the maps from Phaser are rather poor, but there are bits and pieces
which seem to make sense and I was hoping that I could at least
determine the space group.
Any comments are much appreciated!
Sabine
This message has been checked for viruses but the contents of an
attachment may still contain software viruses, which could damage your
computer system: you are advised to perform your own checks. Email
communications with the University of Nottingham may be monitored as
permitted by UK legislation.
