*** For details on how to be removed from this list visit the ***
*** CCP4 home page http://www.ccp4.ac.uk ***
Jay Thompson schrieb:
Hi All,
We have crystals (0.2 x 0.2 x 0.2 mm) that belong to the spacegroup C2,
with a unit cell of 310 A x 290 A x 230 A, 90 102 90 and diffracts
rather poorly at ~4 angstroms. We're trying to collect a Hg and Pt-SAD
datasets, since a MAD dataset is likely not feasible. So far I've
collected data on a couple of crystals (~45 minutes of total exposure
time, Rsym= 0.14, I/sig(I)=7.0, redundancy=4). and I'm having trouble
detecting any anomalous peaks in the Harker sections using the programs
in CCP4. It looks like these crystals are exhibiting radiation decay
(based on unit cell length, scale factor, and mosaicity increases),
after ~10-15 minutes of exposure time on a 2nd generation synchrotron
(like SSRL or ALS). If anyone has any wonderful strategies to collect
SAD data on weakly/poorly diffracting and radiation sensitive crystals,
that would be great!!
I need to decrease the exposure time on the crystal to be able to
collect a complete dataset with some anomalous signal. However, I'm
worried that if I decrease the exposure time the signal-to-noise would
suffer and I would still have trouble finding an anomalous peak in the
Harker section (although redundancy should increase signal-to-noise).
Which is better in terms of improving signal to noise for SAD or MAD
datasets? Would people recommend to keep the exposure time short and
just collect lots of images to increase redundancy and signal-to-noise
or increase exposure time but have a less redundant dataset. So I guess
the question comes down to whether multiple weak reflections is better
than one strong reflections for SAD? Does anyone have any nice
strategies for estimating/optimizing the exposure time for SAD/MAD
datasets? Also do people still like to collect inverse beam for C2
spacegroups? Any comments would be greatly appreciated. Thank you very
much.
JT
Jay,
my advice is: redundancy is most important because it gives you more choices
a) you can (a posteriori) cut your dataset at the most appropriate frame
if raddam is too large (use R_d to find out about the amount of raddam)
b) zero-dose (or rather the more robust quarter-dose) extrapolation
needs redundancy
My rules of thumb are:
- always collect 180° if the space group is tetragonal (unless 422),
trigonal, orthorhombic, monoclinic. >180° if triclinic.
- either (preferred at least for MAD) one pass without (or few)
overloads, or: low-exposure pass first(!), strong-exposure afterwards
- don't go above 0.5° in delta-phi
HTH,
Kay
--
Kay Diederichs http://strucbio.biologie.uni-konstanz.de
email: [EMAIL PROTECTED] Tel +49 7531 88 4049 Fax 3183
Fachbereich Biologie, Universität Konstanz, Box M647, D-78457 Konstanz
