[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[ccp4bb]: Summary:twinning detection
*** For details on how to be removed from this list visit the ***
*** CCP4 home page http://www.ccp4.ac.uk ***
Dear colleagues,
Thank you very much for your suggestions on my question about twinning
detection. Especially, I appreciate the great help of Dr. Eleanor J.
Dodson!
*************************** My original question
****************************
I suspected that my data is twinned and carried out the twinning detection
using the input file detect_twinning.inp in CNS. I got the following
result:
---------------------------averages-over-all-bins-----------------------------
<|I|^2>/(<|I|>)^2 = 4.0639 (2.0 for untwinned, 1.5 for twinned)
(<|F|>)^2/<|F|^2> = 0.5682 (0.785 for untwinned, 0.865 for twinned)
------------------------------------------------------------------------------
Since the statistics of my data is so far away from the idea values for
either twinned or untwinned data, I am not sure if my data is twinned or
not?
(see below for additional informations on my data)
Here are the answers to my question:
Randy Read
>Your data shows even more dispersion from the mean than normal untwinned
>data. This probably indicates that there is serious anisotropy in
>diffraction from your crystal. Take a look at the diffraction patterns
with
>hklview. You should also see that the values are more typical of
untwinned
>data at low resolution, where the effects of anisotropy will be less
>pronounced.
Wim Burmeister
>There must be something seriously wrong with your data.
>- Either you think that you are using F's but in reality you are using
I's
>- Due to the presence of ice or salt you have measurements which have
>aberantly high intensities
>- Your spacegroup is incorrect and does not take into account systematic
>absences for which you measured reflections which are not really there.
>A simililar problem could arise if you get some pseudo-symmetry
corresponding
>to a centered lattice. In this case many of the reflections will be very
weak.
Eleanor J. Dodson
> Have you checked whether your data has a pseudo translational vector?
>( run a native patterson and look for off-origin peaks)
>If so this means many reflections are very weak, and the moments are not
>a useful check.
>
>You may be able to reduce the cell dimensiions, or maybe you have
>centring?
Charlie Bond
>It is possible that your data is not twinned, but anisotropic.
>Anisotropy leads to data which has a wider dynamic range than normal,
>whereas twinning leads to data having a narrower range.
>If you reduce your data with truncate (ccp4i scalepack2mtz etc), it does
>a test for anisotropy.
Gerard DVD Kleywegt
>errr, more than half of your data has I/sigma less than 2 ...
>as keith wilson always says - you have been collecting indices
>rather than data ! maybe you should ask your supervisor for
>some help on data collection and processing ?
Ralf W. Grosse-Kunstleve
>It would be useful to see the Miller indices of the weak reflections. It
>doesn't have to be the whole list, just a couple of hundred should be
>enough.
>Ralf
Eleanor J. Dodson
>what is the fractional patterson vector?
>and what is its relative height to the origin?
Soisson, Stephen Michael
>Your observation that half of the reflections are very weak suggests that
>you have misindexed the data. You should try to rule out this
possibility.
>Looking for trends in the weak data in pseudo precessions (hklview) can
>sometimes be informative, say for instance if you have alternating layers
of
>weak and strong reflections.
Eleanor J. Dodson
>That are certainly very strong off origin Patterson peaks, and I would
>suspect there could be misindexing. It is easy when you have one very
>long cell dimension too But you will have to look back at the images.
>
> Does denzo suggest any other cell with half the cell volume?
********************additional information about the
data*************************
****I am sure that I have not made a mistake in I or F's.
****The reflection images do not show the existence of ice or salt
crystals.
******************************************
********* Index result by DENZO **********
******************************************
Lattice Metric tensor Best cell (symmetrized)
distortion index Best cell (without symmetry restrains)
primitive cubic 77.91% 109.18 154.61 284.06 89.55 89.80 44.88
197.07 197.07 197.07 90.00 90.00 90.00
I centred cubic 38.99% 244.43 303.51 284.06 21.07 90.37 81.06
278.42 278.42 278.42 90.00 90.00 90.00
F centred cubic 40.64% 323.56 322.76 322.35 39.57 140.51 123.04
322.89 322.89 322.89 90.00 90.00 90.00
primitive rhombohedral 6.17% 303.97 303.51 284.06 21.07 21.05 29.39
297.33 297.33 297.33 24.45 24.45 24.45
133.54 133.54 864.90 90.00 90.00 120.00
primitive hexagonal 5.67% 109.10 109.18 284.06 89.80 90.44 90.19
109.14 109.14 284.06 90.00 90.00 120.00
primitive tetragonal 0.25% 109.10 109.18 284.06 89.80 90.44 90.19
109.14 109.14 284.06 90.00 90.00 90.00
I centred tetragonal 4.45% 109.10 109.18 587.61 100.55 79.69 90.19
109.14 109.14 587.61 90.00 90.00 90.00
primitive orthorhombic 0.25% 109.10 109.18 284.06 89.80 90.44 90.19
109.10 109.18 284.06 90.00 90.00 90.00
C centred orthorhombic 0.29% 154.09 154.61 284.06 89.55 90.17 89.96
154.09 154.61 284.06 90.00 90.00 90.00
I centred orthorhombic 4.45% 109.10 109.18 587.61 100.55 79.69 90.19
109.10 109.18 587.61 90.00 90.00 90.00
F centred orthorhombic 4.92% 109.10 244.43 577.68 94.50 79.55 116.70
109.10 244.43 577.68 90.00 90.00 90.00
primitive monoclinic 0.11% 109.10 109.18 284.06 89.80 90.44 90.19
109.10 109.18 284.06 90.00 90.44 90.00
C centred monoclinic 0.10% 154.61 154.09 284.06 90.17 90.45 90.04
154.61 154.09 284.06 90.00 90.45 90.00
primitive triclinic 0.00% 109.10 109.18 284.06 89.80 89.56 89.81
******************************************
********* Result from scalepack **********
******************************************
> Shell I/Sigma in resolution shells:
> Lower Upper No. of reflections with I / Sigma less than
> limit limit 0 1 2 3 5 10 20 >20 total
> 20.00 6.82 1682 2193 2550 2740 2963 3198 3520 1947 5467
> 6.82 5.45 1794 2445 2778 2955 3138 3416 3831 1596 5427
> 5.45 4.77 1919 2441 2721 2876 3072 3344 3781 1628 5409
> 4.77 4.34 1870 2329 2594 2774 2980 3275 3713 1599 5312
> 4.34 4.03 1909 2400 2684 2846 3032 3386 3855 1497 5352
> 4.03 3.79 1916 2473 2799 2994 3226 3560 4039 1341 5380
> 3.79 3.61 1803 2380 2807 3043 3307 3667 4205 1184 5389
> 3.61 3.45 1749 2447 2904 3149 3456 3919 4437 955 5392
> 3.45 3.32 1802 2585 3071 3356 3675 4094 4692 702 5394
> 3.32 3.20 1733 2607 3196 3488 3835 4327 4899 530 5429
> 3.20 3.10 1712 2646 3285 3585 3963 4517 4988 411 5399
> 3.10 3.02 1762 2744 3371 3737 4102 4659 5096 306 5402
> 3.02 2.94 1802 2807 3489 3897 4276 4756 5172 226 5398
> 2.94 2.86 1786 2850 3609 3996 4416 4878 5222 165 5387
> 2.86 2.80 1823 2971 3772 4156 4546 4965 5240 127 5367
>All hkl 27062 38318 45630 49592 53987 59961 66690 14214 80904
>
>
> Shell I/Sigma in resolution shells:
> Lower Upper % of of reflections with I / Sigma less than
> limit limit 0 1 2 3 5 10 20 >20 total
> 20.00 6.82 30.7 40.0 46.5 50.0 54.1 58.4 64.3 35.5 99.8
> 6.82 5.45 33.0 45.0 51.2 54.4 57.8 62.9 70.5 29.4 99.9
> 5.45 4.77 35.3 44.9 50.1 52.9 56.5 61.5 69.6 30.0 99.5
> 4.77 4.34 34.6 43.0 47.9 51.3 55.1 60.5 68.6 29.6 98.2
> 4.34 4.03 35.3 44.4 49.6 52.6 56.0 62.6 71.3 27.7 98.9
> 4.03 3.79 35.3 45.6 51.6 55.2 59.5 65.6 74.5 24.7 99.2
> 3.79 3.61 33.4 44.0 51.9 56.3 61.2 67.8 77.8 21.9 99.7
> 3.61 3.45 32.4 45.3 53.7 58.3 63.9 72.5 82.1 17.7 99.7
> 3.45 3.32 33.4 47.9 56.9 62.2 68.1 75.9 87.0 13.0 100.0
> 3.32 3.20 31.9 48.0 58.8 64.2 70.6 79.7 90.2 9.8 100.0
> 3.20 3.10 31.7 49.0 60.8 66.4 73.3 83.6 92.3 7.6 99.9
> 3.10 3.02 32.6 50.8 62.4 69.2 75.9 86.2 94.3 5.7 100.0
> 3.02 2.94 33.3 51.9 64.6 72.1 79.1 88.0 95.7 4.2 99.9
> 2.94 2.86 33.1 52.8 66.9 74.1 81.9 90.4 96.8 3.1 99.9
> 2.86 2.80 33.9 55.3 70.2 77.3 84.5 92.3 97.5 2.4 99.8
>All hkl 33.3 47.2 56.2 61.1 66.5 73.8 82.1 17.5 99.6
More than half of the reflections are very weak ( < 2 sigma).
******************************************
************** MR solution ***************
******************************************
The spacegroup is P4(3), a=b=109A, c=284A. 4 molecules in ASU.
The 4 moleclues are related by a pseudo 4-fold screw axis at c direction
(a=0, b=0.5, fractional) with a tranlation of 11A (0.039 in fraction).
However, it seems that the screw axis do not extend to the whole unit cell.
******************************************
************ Patterson Vector ************
******************************************
The native patterson map gives the following peaks above 5*sigma in
fractional coordinate.
These peaks can be deduced from the crystallographic 4(3) axis and NCS 4(1)
axis:
GRID 84 84 208
CELL 109.2580 109.2580 282.4230 90.0000 90.0000 90.0000
ATOM Ano 0.0000 0.0000 0.0000 155.53 0.0 BFAC 20.0
ATOM Ano 0.5000 0.5000 0.7118 134.65 0.0 BFAC 20.0
0.7118<->0.75-0.039
ATOM Ano 0.5000 0.5000 0.2882 134.65 0.0 BFAC 20.0
ATOM Ano 0.0000 0.0000 0.4236 87.28 0.0 BFAC 20.0
0.4236<->0.50-0.039*2
ATOM Ano 0.5000 0.5000 0.1357 41.08 0.0 BFAC 20.0
0.1357<->0.25-0.039*3
ATOM Ano 0.5000 0.5000 0.8643 41.08 0.0 BFAC 20.0
ATOM Ano 0.0000 0.0000 0.1509 13.89 0.0 BFAC 20.0
ATOM Ano 0.9791 0.0206 0.0141 5.14 0.0 BFAC 20.0
ATOM Ano 0.0209 0.9794 0.0141 5.14 0.0 BFAC 20.0
ATOM Ano 0.0206 0.0209 0.9859 5.14 0.0 BFAC 20.0
ATOM Ano 0.9791 0.0206 0.9859 5.14 0.0 BFAC 20.0
ATOM Ano 0.9794 0.9791 0.9859 5.14 0.0 BFAC 20.0
ATOM Ano 0.0206 0.0209 0.0141 5.14 0.0 BFAC 20.0
ATOM Ano 0.9794 0.9791 0.0141 5.14 0.0 BFAC 20.0
ATOM Ano 0.0209 0.9794 0.9859 5.14 0.0 BFAC 20.0
*********************************************
************ Diffraction Pattern ************
*********************************************
Using hklview I found that the diffraction intensities of the layers
hk0, hk1, hk2,... show obvious variety, but not in a regular period.
l <I/error>
0 7.2
1 1.5
2 10.6
3 8
4 6.4
5 11
6 0.4
7 12
8 3.2
9 9.5
10 8.9
11 3.9
12 11
13 0.3
14 11.1
15 5.7
16 8.3
17 10.1
18 2.6
19 11.1
20 1.3
21 11.1
22 6.9
23 7.1
24 11.2
25 1.3
26 11.3
27 3.4
28 10.1
29 8.9
30 5.4
31 11.5
32 1.2
33 11.4
34 5
35 9
36 10.3
37 3.1
38 11.9
39 11.6
_________________________________________________________________
与联机的朋友进行交流，请使用 MSN Messenger: http://messenger.msn.com/cn