1) You dont say whether there is a non crystallographic translation 
vector - if that is so the twinning statistics can be misleading. 
SFCHECK analyses this or you can just run a native patterson to 4A say 
and see if there is an off-origin peak.

2) The twinning tests that use the correlation between possibly 
"twinned" reflectons ( ie hkl and kh-l for tetragonal) always say your 
data is twinned if the true symmetry IS 422 when hkl and kh-l are 
symmetry equivalents..  (They should say - either twinned or with a 
higher symmetry..)

3) I believe in the moments plotted in TRUNCATE providing there isnt a 
non crystallographic translation vector..

 Eleanor





Joe Smith wrote:
>   Hi all,
>
> We have collected few X-ray data sets for a protein-RNA complex to
> resolutions of 3.2-3.5A. While processing the data using HKL2000, we
> have obtained following distortion index consistently:
>
>  primitive cubic         19.16%     127.98  74.67 130.85  74.57  85.00  73.91
>                                                111.17 111.17 111.17
> 90.00  90.00  90.00
>
>   I centred cubic         24.26%     132.28 189.90 127.98 120.33  85.97 120.47
>                                                 150.05 150.05 150.05
> 90.00  90.00  90.00
>
>  F centred cubic         22.17%     177.47 190.41 189.90  46.24  90.79  91.90
>                                                  185.93 185.93 185.93
> 90.00  90.00  90.00
>
>  primitive rhombohedral  16.32%  130.85 129.06 174.88 131.21 133.20  86.12
>                                                   144.93 144.93 144.93
> 116.85 116.85 116.85
>                                                   229.30 229.30  74.67
>  90.00  90.00 120.00
>
>  primitive hexagonal     15.40%    130.85 127.98  74.67 106.09  74.57  95.00
>                                                  129.41 129.41  74.67
> 90.00  90.00 120.00
>
>  primitive tetragonal     9.40%       127.98 130.85  74.67  74.57 106.09  95.00
>                                                   129.41 129.41  74.67
>  90.00  90.00  90.00
>
>  I centred tetragonal     0.91%      177.47 174.88  74.67  90.21  88.59  91.47
>                                                  176.18 176.18  74.67
> 90.00  90.00  90.00
>
>  primitive orthorhombic   9.38%     74.67 127.98 130.85  85.00 105.43 106.09
>                                                   74.67 127.98 130.85
> 90.00  90.00  90.00
>
>    C centred orthorhombic   6.53%   74.67 245.95 130.85  89.44 105.43  89.13
>                                                    74.67 245.95 130.85
>  90.00  90.00  90.00
>
>  I centred orthorhombic   0.84%    74.67 174.88 177.47  88.53  91.41  90.21
>                                                   74.67 174.88 177.47
> 90.00  90.00  90.00
>
>  F centred orthorhombic   0.68%    74.67 245.95 252.32  89.16  88.86  89.13
>                                                  74.67 245.95 252.32
> 90.00  90.00  90.00
>
>  primitive monoclinic     6.52%     74.67 130.85 127.98  95.00 106.09  74.57
>                                                   74.67 130.85 127.98
> 90.00 106.09  90.00
>
>  C centred monoclinic     0.49%     74.67 245.95 130.85  89.44 105.43  89.13
>                                                   74.67 245.95 130.85
> 90.00 105.43  90.00
>
>  primitive triclinic           0.00%      74.67 127.98 130.85  85.00
> 74.57  73.91
>
>
> As you see, distortion index table indicates I centered tetragonal, I
> centered orthorhombic, F centered orthorhombic, C centered monoclinic
> and triclinic as possible Bravais lattices.
>
>  Data processed in I centered tetragonal gives low Rmerge in all the
> possible space groups namely I4, I41, I422 and even I4122.  Other
> space groups in lower symmetry lattices also gave low R merge values
> (around 6% in most of the cases).
>
> Since we have not been able to obtain a solution in any of the space
> group from I centered tetragonal to triclinic (I4, I4122, I222, C2 and
> even P1) using Se-SAD, we decided to check the data for any intrinsic
> problem such as twinning.
>
> Cumulative intensity distribution calculated using scalepack2mtz shows
> no sign of twinning. However, data processed in I4 shows nearly a
> perfect twin (twin fraction=0.489 with twin operator 100 0-10 00-1) in
> Yeates server whereas SFcheck indicates a twin fraction of 0.431 with
> twin operator –h,+k,-l. Data processed in I4122, I222, C2 and P1
> doesn't show any twinning due to absence of any twin laws for these
> space groups.
>
> Now my question is:
> -  Are data showing low Rmerge value in I4122 due to nearly perfect
> twin in space group I4?
>
> -  Why cumulative intensity distribution shows a normal pattern for
> the data where as Yeates server and SFcheck indicates nearly a perfect
> twin?  Why Yeates server and SFCheck shows different twin fraction and
> twin operator?
>
> -  Is it possible to detwin this data and use it for structure solution?
>
> Thank you for reading till this line and I am sorry for such a long
> mail. I hope I haven't made any mistake at any stage. I really need
> your valuable suggestions to solve this problem.
>
> Regards
> Joe
>
>
>
>