Re: [ccp4bb]: Twinning/Merging

[Ulrich Baumann <ulrich.baumann@ibc.unibe.ch>]

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

As said before, you should examine each crystal individually, especially
the TRUNCATE output (if available).  Crystals with different twinning
fractions will merge poorly, of course. Also a twinning fraction of
significantly less than 0.5 will give poor statistics for one individual
crystal in the apparent high-symmetry space group. Another very trivial
point: are you sure that you crystals have been indexed consistently? IN
P3(1/2)21 you have two different orientations which will give different
solutions for indexing: h,k,l and -h,-k,l. Each one is perfectly valid but
different crystals must be indexed consistently in the same orientation
before merging. You can keep one crystal as reference indexed as it is and
merge another one indexed as h,k,l or h,-k-l. Normally one would expect
results in the order of 40 % if the crystals are not indexed consistently
but two of your three crystals are of course indexed in the same
orientation and the third one might be indexed in the other orientation by
chance and hence spoil the whole data set. Please not that this indexing
ambiguity has nothing to do with handedness or twinning. Each orientation
is indexed in a right-handed way.



***************************************************
**  Ulrich Baumann, Dpt. f. Chemie und Biochemie **
**  Universitaet Bern, Freiestrasse 3            **
**  CH-3012 Bern, Switzerland                    **
**  phone +41-(0)31-631-4320/4324, fax -4887     **
***************************************************

On Fri, 15 Dec 2000, Daniel Schlieper wrote:

> ***  For details on how to be removed from this list visit the  ***
> ***    CCP4 home page http://www.dl.ac.uk/CCP/CCP4/main.html    ***
> 
> Dear all,
> 
> Of one protein, I used 3 crystals to collect a data set at room
> temperature. The data set is to 96 % complete, diffracts to 2.35 A,
> and has a very high Rmerge value after scalepack: 16 % (77 % in the
> 2.38-2.35 A shell). 
> 
> The twinning test (from Yeates, www.doe-mbi.ucla.edu/Services/Twinning/) 
> gave no twinning at all.
> 
> I used the very same protein to grow crystals of the very same crystal
> form. (The crystals differ in the substrate co-crystallized). These
> crystals were measured at 100 K, and I collected full data sets of
> every crystal. Tested for twinning, it came true, that all of them
> were twins of 5-15% (P3221, (-h,-k,l)).
> 
> Now I wonder if the 3 crystals measured at room temperature have been
> twins as well, and if I may explain the high R-value with that. Is
> this true?
> 
> See you in York!
> 
> Best regards, Daniel
> 
> -- 
> Daniel Schlieper                      Institut fuer Biochemie
>                                       Zuelpicher Strasse 47
> Daniel.Schlieper@Uni-Koeln.De         Universitaet zu Koeln 
> Tel.: +49 221 470-6443, Fax: -5092    50674 Koeln, Germany
> 
> 
> -- 
> Daniel Schlieper                      Institut fuer Biochemie
>                                       Zuelpicher Strasse 47
> Daniel.Schlieper@Uni-Koeln.De         Universitaet zu Koeln 
> Tel.: +49 221 470-6443, Fax: -5092    50674 Koeln, Germany
> 
>