Re: [ccp4bb]: Problems with low mosaicity crystals

[<yxiong@chemistry.ohio-state.edu>]

***  For details on how to be removed from this list visit the  ***
***          CCP4 home page http://www.ccp4.ac.uk         ***

Many thanks again for the quick replies. Let me put together a little summary 
for the disscussion so far and make clear a few things.

First, a few people suggested to check misindexing, beam center position, 
twining ... We are fairly confident these are not the problem, since those 
crystals are well characterized as data had been measured repeatedly before 
and these possibilities were thouroughly checked, and at the time they were 
measureed, there were many other data sets also measured fine (unlikely 
beamline problem or beam centering).

A little more info for the data I can recall:
1. APS 14BMD, Q4; P422(36.2  36.2  74.1); Reso ~1A; Rsym ~10, mos .2-.3.
2. APS 19-ID; SBC-2; same crystal as above; Reso <0.6A; Rsym ~10, mos .21
3. CHESS A1(3/3/02); Q210; #1: P622(143.8 143.8 164.0); Reso ~2.5A; Rsym ~19; 
mos .29. #2: P3121(105.9 105.9 182.1); Reso ~3A; Rsym ~9; mos .27
Note: Rsym's given are for low reso bins. And as said above, on the same trip 
we had data with similiar crystal (different soak/mosaicity) with Rsym's ~4%. 

Regarding the loop vibration, some belive so while some don't. I am inclined 
to think this is the problem.

Jim gave a good illustration of how the vibration (angular vibration) could 
move the reflection in and not the Ewald sphere. Adn Ed showed how the 
vibration of the loop would lead to such an angular movement.

Pete remarked such a vibration would raise the apparent mosaicity (which 
contradict to the low mos measured). This question was partly answer by 
Gerard that the vibration is only in certain orientations thus do not have 
the effect as the mos spread. Also, say, if the angular movement is about 
10~20% of the mos, then we don't really see smeared or more spots, while the 
effect on intensity measured might still be significant (explained below).

Richard and a number of others pointed out "Angular oscillation of the
crystal (several Hz) would be time-averaged over the exposure and have 
similar effect as classic random distribution of microcrystal domains ..."
My thought to this is that the angular oscillation is not homogenious in all 
direction, instead it happens mainly along the cryo stream direction. Thus it 
will affect the spots on the Ewald sphere position perpendicular to this 
direction more (say 80-90% of time staying in diffraction condition), while 
those tangent to this direction might stayed on Ewald sphere all the time! 
(smeared a little probably). 10%-20% of the mos is already significant in 
this regard (easily cause 10% difference in intensity), and it will not show 
in dirraction pattren. And, since the augluar vibration is probably not big, 
only low mossaicity crystals can feel it.

A few people also pointed out spindle problem, beam undulation, inefficiency 
in profile fitting all affect low mosaic crystal more.

Later.

Yong