Re: [ccp4bb]: problems with

["Eleanor J. Dodson" <ccp4@ysbl.york.ac.uk>]

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



  Those SGS are often twinned..

Checks:
1) TRUNCATE plots -
Cumulative intensity plot - is it sinusoidal?
Moments - 2nd moment is often best - is it less than 2.
1.5 indicate  perfect twin.

2) then check native patterson to ~4A. Is there a large offorigin peak 
indicating 2 mols in the same orientation.

3) Check twinning - you can use the Yeats or CNS or the program detwin 
in CCP4.
All will suggest a twin fraction. In these SGS the default twinning 
operator is k h -l  but I guess in P4i that is equivalent to h -k -l

See the twinning notes on the CCP4 documentation for operators..

  Best advice - lok for another crystal!

Eleanor


OnLineHelpForm wrote:
> I am using ccp4 version  release-4_2_2.
> I am using ccp4i.
> My compiler is: native
> I installed using compilesource
> The problem is as follows:
> 
> Hi all!
> 
>    I"ve got data of two crystal forms (P43 and R3) of a
>    monomeric protein. Scaling stats for P43 data are attached
>    below. Resolution in both cases is 2.8A
> 
>    Phasing was performed with amore, unique Molrep was found,
>    2 mol per ASU, showing a pseudo-2-fold. Increasing the
>    symmetry to R32 / P43212 (thats what the adata looked like
>    with hklview) respectivly with one mol per ASU yielded in
>    Rfacs which were about 4-5 0gher. I don't think I'm in
>    the wrong sapcegroup. All other possibilities either scaled
>    badly or I could not get a molrep solution. Refinement
>    stops at Rfree/Rfac of 34/30 in both cases. There is no
>    more defined density in both cases to be seen, while the
>    model contains only about 800f the backbone and 600f
>    the sidechains. 
> 
>    Because of the high disorder and Rfac not coming down I
>    checked for twinning more carefully. i have never observed
>    twinning before, so I can
>    't really tell, if my data is actually twinned. The Wilson
>    stats suggest no twinning, but Iobs is always higher than
>    Icalc (Truncate output same as detect twin from CNS) and a
>    twinning fraction of 0.47 is calculated (see below)
> 
>    Can anyone tell me, what the problem in my two data sets
>    is, what I could/should check, if my data is twinned (and
>    if so what to do next ...).
> 
>    Thanks for any help!!!
> 
>    Tanja
> 
> 
> 
>                            testing for perfect twinning
> 
>     column 1: bin number
>     columns 2: upper resolution limit
>     columns 3: lower resolution limit
>     column 4: number of reflections in bin
>     column 5: average resolution in bin
>     column 6: <|I|^2>/(<|I|>)^2
>     column 7: (<|F|>)^2/<|F|^2>
>     column 8: fraction of theoretically complete data
> 
>      <|I|^2>/(<|I|>)^2 is 2.0 for untwinned data, 1.5 for
>    twinned data
>     (<|F|>)^2/<|F|^2> is 0.785 for untwinned data, 0.865 for
>    twinned data
> 
>     #bin | resolution range | #refl | 
>        1 7.56 30.00 505 10.6717 2.8341
>       0.6892 0.9352
>        2 6.02 7.56 511 6.6736 2.4945
>       0.7310 0.9623
>        3 5.26 6.02 517 5.6053 2.1239
>       0.7711 0.9682
>        4 4.78 5.26 516 5.0055 2.3628
>       0.7349 0.9663
>        5 4.44 4.78 523 4.5986 2.5085
>       0.7198 0.9703
>        6 4.18 4.44 502 4.3052 2.1968
>       0.7573 0.9805
>        7 3.97 4.18 536 4.0735 2.0388
>       0.7794 0.9799
>        8 3.80 3.97 518 3.8785 2.3384
>       0.7337 0.9700
>        9 3.65 3.80 512 3.7245 2.3865
>       0.7534 0.9827
>       10 3.53 3.65 525 3.5858 2.7638
>       0.7289 0.9722
>       11 3.42 3.53 513 3.4702 2.5059
>       0.7324 0.9884
>       12 3.32 3.42 512 3.3680 2.8393
>       0.7219 0.9827
>       13 3.23 3.32 537 3.2741 3.0911
>       0.7120 0.9781
>       14 3.15 3.23 520 3.1923 2.5319
>       0.7343 0.9905
>       15 3.08 3.15 515 3.1149 3.5783
>       0.7277 0.9791
>       16 3.02 3.08 537 3.0477 3.3741
>       0.6909 0.9853
>       17 2.96 3.02 508 2.9848 2.6581
>       0.7305 0.9845
>       18 2.90 2.96 503 2.9281 2.7854
>       0.7168 0.9902
>       19 2.85 2.90 528 2.8743 2.6716
>       0.7387 0.9548
>       20 2.80 2.85 0 0.0000 0.0000
>       0.0000 0.0000
> 
>    ---------------------------averages-over-all-bins-----------------------------
>       <|I|^2>/(<|I|>)^2 = 2.6374 (2.0 for untwinned, 1.5
>      for twinned)
>      (<|F|>)^2/<|F|^2> = 0.7318 (0.785 for untwinned,
>    0.865 for twinned)
>    ------------------------------------------------------------------------------
> 
>    ==============================================================================
> 
>          testing for partial twinning (using statistical
>    method of Yeates)
> 
> 
>    >>>> testing for twinning operator= h,-k,-l
> 
> 
>         <H> = 0.02134: twinning fraction= 0.479 (3669
>    reflections used)
>       <H^2> = 0.00093: twinning fraction= 0.474 (3669
>    reflections used)
> 
>    ==============================================================================
> 
> 
> 
> 
>    scaling P43: 100
>