Re: [ccp4bb] translational NCS & twinning

[Donghyuk Shin]

Dear Eleanor,

Many thanks for you comments.

I have run aimless/pointless with those data sets having unit cell (134/67, 
134/67, 183, 90, 90 120) previously integrated with P31 2 1.
Previously, I forced aimless to not determine laue group, to keep the original 
SG, and now I let aimless determine the SG.

Both aimless and pointless re-indexed both data sets to the P63 2 2 for both 
data sets.
Based on the matthews analysis, it seems impossible to put molecule in to small 
cell (67. 67, 183, 90, 90, 120), and truncate analysis for this large cell 
indicates both tNCS and twinning. I am confused..how to interpret my data sets. 
Does it have both tNCS and twinning? 

For curiosity, I have ran the Phaser by turning on/off the tNCS with larger 
cell (134, 134, 180, 90, 90, 120), and only the phaser 'without tNCS' gave me 
the solution, but still it did not give me the 2 molecules/ASU which should be, 
and just 1mol/ASU.
Again for curiosity, I ran Refmac but results were like below.
-> Refmac without twin: 0.51/0.56 (work/free)
-> Refmac with twin: 0.52/0.59

I am also attached the log file of pointless for both cells.

Going back to the previous post, I am very open to accept that my C2 refinement 
is wrong, happy to learn. But, based on L-test, H-test and dropping R-values 
and model with density, I guess this is quite convincing but as you commented 
maybe I am misleading. Please let me know if you have more comments.


Best wishes,
Donghyuk





To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1
#CCP4I VERSION CCP4Interface 7.0.051
#CCP4I SCRIPT LOG pointless
#CCP4I DATE 15 Jan 2019  13:42:04
#CCP4I USER user
#CCP4I PROJECT donghyuk_d011_x024
#CCP4I JOB_ID 135
#CCP4I SCRATCH /tmp/user
#CCP4I HOSTNAME localhost.localdomain
#CCP4I PID 30946

 
 ###
 ###
 ###
 ### CCP4 7.0.051: POINTLESS version 1.11.8 : 19/12/17##
 ###
 User: user  Run date: 15/ 1/2019 Run time: 13:42:04 


 Please reference: Collaborative Computational Project, Number 4. 2011.
 "Overview of the CCP4 suite and current developments". Acta Cryst. D67, 235-242.
 as well as any specific reference in the program write-up.

 Input command lines 

XDSIN /home/user/Donghyuk/d011_x024/process_donghyuk/xds_016/XDS_ASCII.HKL
HKLOUT /home/user/Donghyuk/d011_x024/ccp4/XDS_016_pointless.mtz
## This script run with the command   ##
# /home/user/Downloads/destination/ccp4-7.0/bin/pointless


 End of input

Release Date: 19th December 2017


**
**
* POINTLESS  *
*   1.11.8   *
**
*   Determine Laue group from unmerged intensities   *
* Phil Evans MRC LMB, Cambridge  *
* Uses cctbx routines by Ralf Grosse-Kunstleve et al.*
**
**


Reading XDS ascii file from file /home/user/Donghyuk/d011_x024/process_donghyuk/xds_016/XDS_ASCII.HKL

Header lines:

!FORMAT=XDS_ASCIIMERGE=FALSEFRIEDEL'S_LAW=FALSE
!OUTPUT_FILE=XDS_ASCII.HKLDATE=11-Jan-2019
!Generated by CORRECT   (VERSION Jan 26, 2018  BUILT=20180808)
!PROFILE_FITTING= TRUE 
!NAME_TEMPLATE_OF_DATA_FRAMES=/home/user/Donghyuk/d011_x024/x024_C11_1_??.h5 GENERIC
!DATA_RANGE=   1 900
!ROTATION_AXIS=  0.99  0.001086 -0.000195
!OSCILLATION_RANGE=  0.15
!STARTING_ANGLE= 0.000
!STARTING_FRAME=   1
!INCLUDE_RESOLUTION_RANGE=50.000 1.582
!SPACE_GROUP_NUMBER=  152
!UNIT_CELL_CONSTANTS=   134.222   134.222   182.665  90.000  90.000 120.000
!UNIT_CELL_A-AXIS=-9.656   -13.195   133.223
!UNIT_CELL_B-AXIS=36.958   117.512   -53.297
!UNIT_CELL_C-AXIS=  -175.05651.620-7.575
!REFLECTING_RANGE_E.S.D.= 0.163
!BEAM_DIVERGENCE_E.S.D.= 0.029
!X-RAY_WAVELENGTH=  0.87
!INCIDENT_BEAM_DIRECTION= -0.002009 -0.001292  1.10
!FRACTION_OF_POLARIZATION=   0.990
!POLARIZATION_PLANE_NORMAL=  0.00  1.00  0.00
!AIR=  0.000339
!SILICON=  3.942720
!SENSOR_THICKNESS=  0.45
!DETECTOR=EIGER 
!OVERLOAD=   300
!NX=  4150  NY=  4371QX=  0.075000  QY=  0.075000
!ORGX=   2066.92  ORGY=   2186.64
!DETECTOR_DISTANCE=   260.894
!DIRECTION_OF_DETECTOR_X-AXIS=   1.0   0.0   0.0
!DIRECTION_OF_DETECTOR_Y-AXIS=   

Re: [ccp4bb] translational NCS & twinning

[Eleanor Dodson]

Donghyuk - testing your "twinning" in a lower symmetry space group can be
misleading.
Many checks look to see if twin related reflections are similar.
But the twin laws are often the same as the symmetry operators, so if you
have ignored the true symmetry, you will wrongly assume you have twinning.

In this case I would do the following.
Look at the small trigonal cell in pointless/aimless etc.  (134.223/2
134.223/2 182.666, 90, 90, 120)
(You can just input the data integrated in C2 I think and pointless will
check alternate cells.
Your C2  data with these cell dimensions can be reindexed into a trigonal
cell
67.2  67.2 182.9  90.0  90.0 120.0 )

Then pointless will check the likely pointgroup symmetrys -  P3 ?  P3/mmm
?  P6?  etc and give you a score for each symmetry operator.

And since you now have no non-cryst translation you can believe the twin
tests much more.

OK?
Eleanor









On Tue, 15 Jan 2019 at 10:57, Donghyuk Shin  wrote:

> Dear all,
>
> Thank you very much for all of your constructive comments.
> Based on all of your comments, I have done several treatment on my data
> sets, and here are some questions related with both your comments and my
> results.
>
> First of all, I would like to make a short note summarizing all of your
> comments for the people who will have similar problem.
>
> 1) Twinning and tNCS has opposite effects to each other, and one should
> carefully analyze the data if one of them present.
> 2) Simple tNCS can be automatically analyzed and corrected during Phaser,
> but if it does not work, you may try to lower symmetry, decrease the size
> of the Uni cell, or turning off the tNCS option during phaser.
> 3) If one think there is twinning in the crystal, to make sure whether
> your data is twinned, it is better to take a look twinning test than seeing
> the R-values from refinement with twin operators. Applying twin operators
> always give you better R- values.
>
> Then, here are some questions related to my current analysis.
> Q1. As many of you concerned and I also had speculation on my C2 cell
> refinement with twin operators, I tried to analyze twinning of my datasets.
> (ATTACHED)
> As you can see the log file from truncate, I guess my crystal is twinned.
> In addition to this, I also followed the discussion between Randy and Lan.
> I wanted to make sure whether my refinement with twin operators is correct
> or not.  As Randy recommend and based on the paper from Garib N. Murshudov (
> http://www.ysbl.york.ac.uk/refmac/papers/Rfactor.pdf), I could see the
> drop of R-values with twin operators is always possible, however, in this
> paper, I could see the gap between refinements w/ or w/o twin operators is
> quite smaller than my case. (0.49-> 0.41 or 0.58 -> 0.52 vs my case 0.39 ->
> 0.23). Together, based on both twinning-test and good R-values, I now
> believe C2 refinement with twin operators are true. What do you think?
>
> Q2-1. Because some of you asked me about the original spots and
> re-indexing the data sets with high symmetry SG,  I went back to mosflm,
> and here are some images from my analysis. As you can clearly see there are
> strong/weak spots in the frames, and if you increase the threshold you can
> pick the strong spots (ATTACHED). Based on this, initially, I assumed that
> my crystal has tNCS, because I thought this strong - weak pattern is caused
> by tNCS. or Am I wrong?
>
> Q2-2. During new indexing, I choose two unit cells (large and small) and
> integrated them into 2 SG (P31 1 2 or P31 2 1) as Lijun commented. (four
> data sets)
> All of these data sets indicated twinning, while smaller one showed higher
> twinning fractions. Then, I ran phaser, and found that phaser only found
> the solution from P31 2 1SG from both large and small cells. However, I
> could see crash of molecules from large cell, and I decided to stick with
> small cell again. Then, during refinement with twin operators the R-values
> drops from (0.46/0.49) to (0.37/0.42), even the map looks already good. I
> assume that applying high symmetry SG is not possible for this case, and
> again C2 refinement might be correct.
>
> Q2-3. Related to Q1 and 2-1, Now, I think C2 (small cell) refinement with
> twin operator is right solution. Then, here I have another question.
> Based on the observations from frames, I guessed my crystal has tNCS, and
> I only followed strong spots. (Maybe I am wrong ?)
> Then, data sets with only strong spots suggests my crystal is twinned.
> Together, I am thinking my crystal has both tNCS and twinning, and I could
> solve the structure by following strong spots with twin operators.
>
> Again, I would like to express my special thanks to you all for giving me
> valuable comments.
> I hope this post will also useful for others in the future.
>
> Donghyuk
>
>
> 
>
> To unsubscribe from the CCP4BB list, click the following link:
> 

Re: [ccp4bb] translational NCS & twinning

[Guan, Lan]

Dear Randy,
Thank you for the important information and the useful link.  I have another 
question to bother you.
How accurately can these twining assessment programs report the data?  What 
could prevent these programs from reporting twinning data? (Twinned data are 
reported as untwinned data)

 Thanks,

Lan




On Jan 12, 2019, at 6:19 AM, Randy Read 
mailto:rj...@cam.ac.uk>> wrote:

CAUTION: This email originated from outside of TTUHSC. Do not click links or 
open attachments unless you recognize the sender and know the content is safe.


Dear Lan,

Yes, that’s a serious problem that has led some people astray, including a few 
papers where people got apparently good R-factors by invoking non-existent 
twinning.

You can find a brief discussion of this point on the CCP4 wiki 
(https://urldefense.proofpoint.com/v2/url?u=https-3A__strucbio.biologie.uni-2Dkonstanz.de_ccp4wiki_index.php_R-2Dfactors=DwIFaQ=dIAUvHjI5lMnDD45iB3vgA=SZfWxbou0cinb5MH936uQKMweCFFe1qb2Aj8yA2JJ_E=Izy3-mP_RaC3MwlNJdICSKPXfwcW_MFeFmkKfem5tEo=ccGuy7JKlEPfv7NvBSM2aRt4EdZz6m-oEwgKBv9Bl4k=),
 which has links to a couple of publications where Garib Murshudov has worked 
out what happens to R-factors when you treat the data as twinned even when it 
isn’t.

Best wishes,

Randy

-
Randy J. Read
Department of Haematology, University of Cambridge
Cambridge Institute for Medical ResearchTel: +44 1223 336500
Wellcome Trust/MRC Building Fax: +44 1223 336827
Hills RoadE-mail: 
rj...@cam.ac.uk
Cambridge CB2 0XY, U.K.   
www-structmed.cimr.cam.ac.uk

On 11 Jan 2019, at 23:05, Guan, Lan 
mailto:lan.g...@ttuhsc.edu>> wrote:

Hi Randy,


As Jacob and others have mentioned, you will always get lower R-factors once 
you treat the data as being twinned, and the more twin operators the bigger the 
reduction in R-factors.

Do normal data with no twinning, but refined with twin operator(s), show 
similar phenomenon?  If it decreases R-factors for normal data, how much it can 
achieve (5-10% lower)?

Thanks,


Lan



So you need very strong evidence, independent of R-factors, to invoke twinning. 
 In this case, the L-test should be reasonably trustworthy even in the presence 
of tNCS, and your L-test values are close to what one would expect for an 
untwinned crystal.  At most you have partial twinning, or perhaps twinning of a 
pseudo-symmetric crystal (a possibility Phil mentioned), where the effects on 
intensity statistics are reduced.

One way to address a problem like this is to solve the structure in a lower 
symmetry space group (as you have done), but then to check whether the MR 
solution actually obeys the higher symmetry.  You can do this by looking at the 
crystal packing or at merging statistics for Fcalcs after a refinement with 
highly restrained NCS.  A similar sort of analysis is automated in the Zanuda 
tool in CCP4.

Dealing with potential complications from combinations of twinning and 
pseudosymmetry is one of the more challenging aspects of crystallography, but 
it's a good learning experience.  Good luck!

Randy Read

On 10 Jan 2019, at 22:38, Donghyuk Shin 
mailto:sdh...@gmail.com>> wrote:

Dear all,

Thank you very much for all of your suggestions and sharing experiences.
As many of you commented, the current small unit cell C2 refinement seems to be 
incorrect or correct, and I should put some efforts to crack this question.

- To Phill Jeffrey,
The idea, trying to find high symmetry SG with small unit cell C2 data is good 
idea, and I will try this.
For your last comments, identifiable electron density differences between each 
chain,
I guess there should not be other densities between chains if my current SG and 
model is correct. Am I right?

- To Ethan,
Turning off the automatic_tNCS_option seems to be good option.
I think, my current data seems to be twinned then tNCS which I am not sure at 
this moment. But I will keep your advice in my mind.

- To Phoebe A. Rice,
It is quite interesting that you also could get structure solution by indexing 
strong spots and having smaller unit cell.
Actually, I was wondering how it was possible that having half-sized unit cell 
could have solution, while full-sized unit cell could not.
It will be great if you can share your experience a bit more (e.g the size of 
smaller unit cell used in initial search for both 1szp and 3pkz)

Again, thank you very much for all of your suggestion.

Best wishes,
Donghyuk



To unsubscribe from the CCP4BB list, click the following link:

Re: [ccp4bb] translational NCS & twinning

[Randy Read]

Dear Lan,

Yes, that’s a serious problem that has led some people astray, including a few 
papers where people got apparently good R-factors by invoking non-existent 
twinning.

You can find a brief discussion of this point on the CCP4 wiki 
(https://strucbio.biologie.uni-konstanz.de/ccp4wiki/index.php/R-factors), which 
has links to a couple of publications where Garib Murshudov has worked out what 
happens to R-factors when you treat the data as twinned even when it isn’t.

Best wishes,

Randy

-
Randy J. Read
Department of Haematology, University of Cambridge
Cambridge Institute for Medical ResearchTel: +44 1223 336500
Wellcome Trust/MRC Building Fax: +44 1223 336827
Hills RoadE-mail: 
rj...@cam.ac.uk
Cambridge CB2 0XY, U.K.   
www-structmed.cimr.cam.ac.uk

> On 11 Jan 2019, at 23:05, Guan, Lan  wrote:
> 
> Hi Randy,
> 
> 
>> As Jacob and others have mentioned, you will always get lower R-factors once 
>> you treat the data as being twinned, and the more twin operators the bigger 
>> the reduction in R-factors.  
> 
> Do normal data with no twinning, but refined with twin operator(s), show 
> similar phenomenon?  If it decreases R-factors for normal data, how much it 
> can achieve (5-10% lower)?
> 
> Thanks,
> 
> 
> Lan
> 
> 
> 
>> So you need very strong evidence, independent of R-factors, to invoke 
>> twinning.  In this case, the L-test should be reasonably trustworthy even in 
>> the presence of tNCS, and your L-test values are close to what one would 
>> expect for an untwinned crystal.  At most you have partial twinning, or 
>> perhaps twinning of a pseudo-symmetric crystal (a possibility Phil 
>> mentioned), where the effects on intensity statistics are reduced.
>> 
>> One way to address a problem like this is to solve the structure in a lower 
>> symmetry space group (as you have done), but then to check whether the MR 
>> solution actually obeys the higher symmetry.  You can do this by looking at 
>> the crystal packing or at merging statistics for Fcalcs after a refinement 
>> with highly restrained NCS.  A similar sort of analysis is automated in the 
>> Zanuda tool in CCP4.
>> 
>> Dealing with potential complications from combinations of twinning and 
>> pseudosymmetry is one of the more challenging aspects of crystallography, 
>> but it's a good learning experience.  Good luck!
>> 
>> Randy Read
>> 
>>> On 10 Jan 2019, at 22:38, Donghyuk Shin  wrote:
>>> 
>>> Dear all,
>>> 
>>> Thank you very much for all of your suggestions and sharing experiences.
>>> As many of you commented, the current small unit cell C2 refinement seems 
>>> to be incorrect or correct, and I should put some efforts to crack this 
>>> question.
>>> 
>>> - To Phill Jeffrey,
>>> The idea, trying to find high symmetry SG with small unit cell C2 data is 
>>> good idea, and I will try this.
>>> For your last comments, identifiable electron density differences between 
>>> each chain,
>>> I guess there should not be other densities between chains if my current SG 
>>> and model is correct. Am I right?
>>> 
>>> - To Ethan,
>>> Turning off the automatic_tNCS_option seems to be good option.
>>> I think, my current data seems to be twinned then tNCS which I am not sure 
>>> at this moment. But I will keep your advice in my mind.
>>> 
>>> - To Phoebe A. Rice,
>>> It is quite interesting that you also could get structure solution by 
>>> indexing strong spots and having smaller unit cell.
>>> Actually, I was wondering how it was possible that having half-sized unit 
>>> cell could have solution, while full-sized unit cell could not.
>>> It will be great if you can share your experience a bit more (e.g the size 
>>> of smaller unit cell used in initial search for both 1szp and 3pkz)
>>> 
>>> Again, thank you very much for all of your suggestion.
>>> 
>>> Best wishes,
>>> Donghyuk
>>> 
>>> 
>>> 
>>> To unsubscribe from the CCP4BB list, click the following link:
>>> https://urldefense.proofpoint.com/v2/url?u=https-3A__www.jiscmail.ac.uk_cgi-2Dbin_webadmin-3FSUBED1-3DCCP4BB-26A-3D1=DwIFAg=dIAUvHjI5lMnDD45iB3vgA=SZfWxbou0cinb5MH936uQKMweCFFe1qb2Aj8yA2JJ_E=1tRkoB-BHhBWDCtbPikRHJfYIDIqpaGbVoF5xFjTLD4=007v9D7bqs0h2uVFwau-m4cmT4PVlWyYZ2UVxJvWkrs=
>> 
>> --
>> Randy J. Read
>> Department of Haematology, University of Cambridge
>> Cambridge Institute for Medical Research  Tel: + 44 1223 336500
>> Wellcome Trust/MRC Building   Fax: + 44 1223 336827
>> Hills RoadE-mail: rj...@cam.ac.uk
>> Cambridge CB2 0XY, U.K.   www-structmed.cimr.cam.ac.uk
>> 
>> 
>> 
>> To unsubscribe from the CCP4BB list, click the following link:
>> 

Re: [ccp4bb] translational NCS & twinning

[Guan, Lan]

Hi Randy,


> As Jacob and others have mentioned, you will always get lower R-factors once 
> you treat the data as being twinned, and the more twin operators the bigger 
> the reduction in R-factors.  

Do normal data with no twinning, but refined with twin operator(s), show 
similar phenomenon?  If it decreases R-factors for normal data, how much it can 
achieve (5-10% lower)?

Thanks,


Lan



> So you need very strong evidence, independent of R-factors, to invoke 
> twinning.  In this case, the L-test should be reasonably trustworthy even in 
> the presence of tNCS, and your L-test values are close to what one would 
> expect for an untwinned crystal.  At most you have partial twinning, or 
> perhaps twinning of a pseudo-symmetric crystal (a possibility Phil 
> mentioned), where the effects on intensity statistics are reduced.
> 
> One way to address a problem like this is to solve the structure in a lower 
> symmetry space group (as you have done), but then to check whether the MR 
> solution actually obeys the higher symmetry.  You can do this by looking at 
> the crystal packing or at merging statistics for Fcalcs after a refinement 
> with highly restrained NCS.  A similar sort of analysis is automated in the 
> Zanuda tool in CCP4.
> 
> Dealing with potential complications from combinations of twinning and 
> pseudosymmetry is one of the more challenging aspects of crystallography, but 
> it's a good learning experience.  Good luck!
> 
> Randy Read
> 
>> On 10 Jan 2019, at 22:38, Donghyuk Shin  wrote:
>> 
>> Dear all,
>> 
>> Thank you very much for all of your suggestions and sharing experiences.
>> As many of you commented, the current small unit cell C2 refinement seems to 
>> be incorrect or correct, and I should put some efforts to crack this 
>> question.
>> 
>> - To Phill Jeffrey,
>> The idea, trying to find high symmetry SG with small unit cell C2 data is 
>> good idea, and I will try this.
>> For your last comments, identifiable electron density differences between 
>> each chain,
>> I guess there should not be other densities between chains if my current SG 
>> and model is correct. Am I right?
>> 
>> - To Ethan,
>> Turning off the automatic_tNCS_option seems to be good option.
>> I think, my current data seems to be twinned then tNCS which I am not sure 
>> at this moment. But I will keep your advice in my mind.
>> 
>> - To Phoebe A. Rice,
>> It is quite interesting that you also could get structure solution by 
>> indexing strong spots and having smaller unit cell.
>> Actually, I was wondering how it was possible that having half-sized unit 
>> cell could have solution, while full-sized unit cell could not.
>> It will be great if you can share your experience a bit more (e.g the size 
>> of smaller unit cell used in initial search for both 1szp and 3pkz)
>> 
>> Again, thank you very much for all of your suggestion.
>> 
>> Best wishes,
>> Donghyuk
>> 
>> 
>> 
>> To unsubscribe from the CCP4BB list, click the following link:
>> https://urldefense.proofpoint.com/v2/url?u=https-3A__www.jiscmail.ac.uk_cgi-2Dbin_webadmin-3FSUBED1-3DCCP4BB-26A-3D1=DwIFAg=dIAUvHjI5lMnDD45iB3vgA=SZfWxbou0cinb5MH936uQKMweCFFe1qb2Aj8yA2JJ_E=1tRkoB-BHhBWDCtbPikRHJfYIDIqpaGbVoF5xFjTLD4=007v9D7bqs0h2uVFwau-m4cmT4PVlWyYZ2UVxJvWkrs=
> 
> --
> Randy J. Read
> Department of Haematology, University of Cambridge
> Cambridge Institute for Medical Research  Tel: + 44 1223 336500
> Wellcome Trust/MRC Building   Fax: + 44 1223 336827
> Hills RoadE-mail: rj...@cam.ac.uk
> Cambridge CB2 0XY, U.K.   www-structmed.cimr.cam.ac.uk
> 
> 
> 
> To unsubscribe from the CCP4BB list, click the following link:
> https://urldefense.proofpoint.com/v2/url?u=https-3A__www.jiscmail.ac.uk_cgi-2Dbin_webadmin-3FSUBED1-3DCCP4BB-26A-3D1=DwIFAg=dIAUvHjI5lMnDD45iB3vgA=SZfWxbou0cinb5MH936uQKMweCFFe1qb2Aj8yA2JJ_E=1tRkoB-BHhBWDCtbPikRHJfYIDIqpaGbVoF5xFjTLD4=007v9D7bqs0h2uVFwau-m4cmT4PVlWyYZ2UVxJvWkrs=




To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1


smime.p7s
Description: S/MIME cryptographic signature

Re: [ccp4bb] translational NCS & twinning

[lijunliuks]

Donghyuk:

How good were your original images?  If the paeks are sharp and clean, your processing could be easier; or you need more efforts.  The data processing statistics is more important for troubleshooting.

Your showed an apparent SG of p6322, but it does have to be exact, especially when with various NCS.  XTRIAGE output the list of 6(1) related systematic absences but none of them went above 10 I/sig(I),  althought not atypical, have you checked the I's against their neighboring I(0 0 6n) to confirm?  Did you try all possible P6i22 within your phaser search?

With twinning test, the closest low symmetry test would be for P6i and P3i21   and P3i12, the later 2 you seemed to have skipped?  High odered twinning in P3i could be possible but would be hard to deal with.

A C2 lattice could be reindexed to a new one with halved a and halved b, which is possible, but if the new one were still with C2, then it would be cautious.

Lijun Liu
DLX Scientific




发自我的小米手机在 Donghyuk Shin ，2019年1月10日 上午4:22写道：Dear all,

I am having tough time with my Xtal data sets those seem to be twinned or have translational NCS, and it will be greatly appreciated if you can give me some advices or comments!

Data was initially processed with XDS and scaled with aimless without specifying certain space group (SG).
Aimless picked the P 63 2 2 for the best SG, but the xtriage indicates there is non-origin peak after patterson analysis. (attached log)
And, I could not get the proper MR-solution from this data sets.

Because I read that twinning and tNCS cannot be properly distinguished at high SG, I went down to subgroup either P32 or P6 assuming that there is twinning which make data set seems to have apparently high SG. (procedure was same XDS->aimless but I specified the SG to keep them)
Then, xtriage still indicates there is non-origin peak as before, but found twin laws for the data sets (attached log).
However, I still could not get the right MR-solution.
Then, I went even further down to P3 or C2, and xtriage found more twin laws which is make sense because of the lower SG. (attached log)
Again, I could not get the MR-solution.
For all the MR running above, I assumed that phaser(ccp4 module) automatically applied tNCS if they present. or should I have to tick on button in the expert parameters?

Then, I went back to the image and processed the datasets with mosflm by checking the indexed spots.
During this step, I played with the threshold for indexing to follow the strong spot for get correct SG.
I am not sure whether this is correct or not, but by putting high threshold for indexing (e.g. ~15) I could index the data with C2 which has half dimension for a,b axes (116.348,  67.218, 182.861,  90, 90, 90) to the original unit cell (232.533, 134.202, 182.67, 90, 90, 90).
With this, I could put 3 molecules in ASU by MR. During refinement, I felt that the R values were not dropping, and I applied twin refinement.
without twin refinement the R values were (0.39/0.44, work/free), and applying twin refinement gave me significantly better values (0.23/0.26).
Because there were 6 twin operators which may cause this huge R value drop, I speculate whether this is true or not.

Your comments will be greatly helpful! 

With you all the best,
Donghyuk





To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1




To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1

[ccp4bb] AW: [ccp4bb] AW: [ccp4bb] AW: [EXTERNAL] [ccp4bb] translational NCS & twinning

[IGBMC]

Le Vendredi 11 Janvier 2019 12:18 CET,  a écrit:

Thank you Herman you for this clear and informative comments !!
Particularly your last point, which I found very instructive.
Now I remember the title of the article by T.O. Yeates "Protein crystals and 
their evil twins" in 1999. And this corresponds well to the difficulty of the 
problem.
Philippe D.

Philippe Dumas

> Dear Philippe,
>
> As Randy just pointed out, when twinning, pseudosymmetry and other 
> pathologies come into play, things really get complicated.
> I agree with what you said but for the current problem, things may be more 
> complicated.
>
> To summarize:
> - "bona fide" twinning: there are two different, intergrown crystals and the 
> intensities of both crystals just add up.
> - "nano scale" twinning, called statistical disorder: the two orientations 
> are randomly distributed through the crystals and the diffraction of both 
> conformations interferes. The conformations behave like alternate 
> conformations.
> - Twinning/pseudosymmetry/wrong unit cell etc.: Here the two conformations, 
> present in the large (true) unit cell and related by crystallographic or 
> noncrystallographic symmetry, may not fit in the small (false) unit cell and 
> be accounted for by introducing twinning where none is present. Especially 
> with 6 twinning operators, the refinement programs have a lot of room to 
> tweak around to reduce the R-factors.
>
> Therefore my advice would be:
> - Critically check the space group and especially how weak are the weak 
> reflections discarded with the small unit cell?
> - the solution you got in the small unit cell may be a subset of what is 
> present in the large unit cell, so I would also try molecular replacement 
> with the ensemble of molecules you got in the small unit cell.
>
> My two cents,
> Herman
>
>
> -Ursprüngliche Nachricht-
> Von: DUMAS Philippe (IGBMC) [mailto:p.du...@ibmc-cnrs.unistra.fr]
> Gesendet: Freitag, 11. Januar 2019 11:52
> An: Schreuder, Herman /DE
> Cc: CCP4BB@JISCMAIL.AC.UK
> Betreff: Re: [ccp4bb] AW: [ccp4bb] AW: [EXTERNAL] [ccp4bb] translational NCS 
> & twinning
>
>
> Le Vendredi 11 Janvier 2019 09:07 CET, herman.schreu...@sanofi.com a écrit:
>
> Dear Herman,
> As far as I understood the twinning problem, what you say is only true in 
> some occasions, and not in others.
> If the "macroscopic" domains are so small that they are smaller than the 
> X-rays coherence length, then you may do what you say because the X-rays 
> emitted by the different domains can interfere.
> But, if the domains become large, the X-rays emitted by the different domains 
> do not interfere anymore and you have to add the weighted intensities, not 
> the amplitudes, of each domain.
> I hope I understood well your comment and did not "interfere negatively" with 
> the thread...
> Best
> Philippe Dumas
>
> > Dear Lan,
> >
> > Thank you for your compliment. I do not use Xtriage, so I did not bother 
> > looking at the log files.
> >
> > What I meant to say is that with twinning, the crystal has different 
> > macroscopic domains where the molecules have different orientations, say 
> > one domain with orientation A and one domain with orientation B. Since 
> > these domains grow on top of each other, they are usually related by a twin 
> > operator similar to a crystallographic operator such as a twofold axis.
> > The fourier transform of the electron density of the crystal is the 
> > convolution of the fourier transform of the individual molecules with the 
> > crystal lattice, with the fourier transform of the individual molecules 
> > usually giving the stronger contribution. So to get a solution with a 
> > decent R-factor, one must include all orientations (A, B etc.) in the 
> > model, with the position of the molecules in the crystal lattice 
> > contributing less to the diffraction pattern. So one can put the 
> > orientations on top of each other in a small unit cell using twinning, or 
> > put them in a larger unit cell at different positions using 
> > crystallographic or non-crystallographic symmetry. That is what I meant be 
> > "twinning" (N)CS.
> >
> > Hope this makes my remark a little clearer.
> >
> > Best,
> > Herman
> >
> > PS: While other BB readers may have had the same question, I have posted 
> > the reply to the BB. I hope you don't mind.
> >
> >
> > -Ursprüngliche Nachricht-
> > Von: Guan, Lan [mailto:lan.g...@ttuhsc.edu]
> > Gesendet: Donnerstag, 10. Januar 2019 20:53
> > An: Schreuder, Herman /DE
> > Betreff: 

[ccp4bb] AW: [ccp4bb] AW: [ccp4bb] AW: [EXTERNAL] [ccp4bb] translational NCS & twinning

[Herman . Schreuder]

Dear Philippe,
 
As Randy just pointed out, when twinning, pseudosymmetry and other pathologies 
come into play, things really get complicated.
I agree with what you said but for the current problem, things may be more 
complicated. 

To summarize:
- "bona fide" twinning: there are two different, intergrown crystals and the 
intensities of both crystals just add up.
- "nano scale" twinning, called statistical disorder: the two orientations are 
randomly distributed through the crystals and the diffraction of both 
conformations interferes. The conformations behave like alternate conformations.
- Twinning/pseudosymmetry/wrong unit cell etc.: Here the two conformations, 
present in the large (true) unit cell and related by crystallographic or 
noncrystallographic symmetry, may not fit in the small (false) unit cell and be 
accounted for by introducing twinning where none is present. Especially with 6 
twinning operators, the refinement programs have a lot of room to tweak around 
to reduce the R-factors.

Therefore my advice would be:
- Critically check the space group and especially how weak are the weak 
reflections discarded with the small unit cell?
- the solution you got in the small unit cell may be a subset of what is 
present in the large unit cell, so I would also try molecular replacement with 
the ensemble of molecules you got in the small unit cell.

My two cents,
Herman


-Ursprüngliche Nachricht-
Von: DUMAS Philippe (IGBMC) [mailto:p.du...@ibmc-cnrs.unistra.fr] 
Gesendet: Freitag, 11. Januar 2019 11:52
An: Schreuder, Herman /DE
Cc: CCP4BB@JISCMAIL.AC.UK
Betreff: Re: [ccp4bb] AW: [ccp4bb] AW: [EXTERNAL] [ccp4bb] translational NCS & 
twinning


Le Vendredi 11 Janvier 2019 09:07 CET, herman.schreu...@sanofi.com a écrit:

Dear Herman,
As far as I understood the twinning problem, what you say is only true in some 
occasions, and not in others.
If the "macroscopic" domains are so small that they are smaller than the X-rays 
coherence length, then you may do what you say because the X-rays emitted by 
the different domains can interfere.
But, if the domains become large, the X-rays emitted by the different domains 
do not interfere anymore and you have to add the weighted intensities, not the 
amplitudes, of each domain.
I hope I understood well your comment and did not "interfere negatively" with 
the thread...
Best
Philippe Dumas

> Dear Lan,
>
> Thank you for your compliment. I do not use Xtriage, so I did not bother 
> looking at the log files.
>
> What I meant to say is that with twinning, the crystal has different 
> macroscopic domains where the molecules have different orientations, say one 
> domain with orientation A and one domain with orientation B. Since these 
> domains grow on top of each other, they are usually related by a twin 
> operator similar to a crystallographic operator such as a twofold axis.
> The fourier transform of the electron density of the crystal is the 
> convolution of the fourier transform of the individual molecules with the 
> crystal lattice, with the fourier transform of the individual molecules 
> usually giving the stronger contribution. So to get a solution with a decent 
> R-factor, one must include all orientations (A, B etc.) in the model, with 
> the position of the molecules in the crystal lattice contributing less to the 
> diffraction pattern. So one can put the orientations on top of each other in 
> a small unit cell using twinning, or put them in a larger unit cell at 
> different positions using crystallographic or non-crystallographic symmetry. 
> That is what I meant be "twinning" (N)CS.
>
> Hope this makes my remark a little clearer.
>
> Best,
> Herman
>
> PS: While other BB readers may have had the same question, I have posted the 
> reply to the BB. I hope you don't mind.
>
>
> -Ursprüngliche Nachricht-
> Von: Guan, Lan [mailto:lan.g...@ttuhsc.edu]
> Gesendet: Donnerstag, 10. Januar 2019 20:53
> An: Schreuder, Herman /DE
> Betreff: Re: [ccp4bb] AW: [EXTERNAL] [ccp4bb] translational NCS & twinning
>
> Dear Herman,
> I have read your insightful comments on twining and tNCS for years now, which 
> is very useful and helpful!  Thanks,
>
> For Donghyuk’s case, do you think that he really has a twinning issue?  With 
> a lower symmetry, all possible twining operators is always reported in the 
> Xtriage and did not mean a real twin existed.  His L test shows a twin 
> fraction of 0.00 in his log file.  The intensity statistics does not really 
> indicate an actually twinning.  Base on the refinement, twin law is needed to 
> get refinement going.  It looks like a twin.  I am confused...
>
> > the molecules are related by "twinning" (N)CS?
>
>
> What does this mean “ twinning (N)CS"?  Would you p

Re: [ccp4bb] translational NCS & twinning

[Eleanor Dodson]

That P6322 Xtriage shows pretty clearly that you have good data, and that
it is not twinned..

So I think you stick with that  spacegroup - introducing false twin laws to
explain true crystal symmetry reduces the R values at the expense of
getting clear electron density.

You could process the data in a cell with the a b axes halved, choosing
only the strong reflections not affected by the non-cryst Patterson vector,
but in fact the MR programs take such a translation into account during
structure solution .

Eleanor Dodson




On Fri, 11 Jan 2019 at 10:54, Randy Read  wrote:

> Hi,
>
> I'm just catching up on the BB after the CCP4 Study Weekend!
>
> Going back to the beginning of this thread, it's true that tNCS and
> twinning have opposite effects on the intensity moments, which can mask
> each other.  However, for simple cases (as this one appears to be) with a
> single NCS translation, the tNCS analysis in Phaser seems to be pretty
> successful at deconvoluting those effects.  It would be interesting to see
> the overall second moments and the second moments as a function of
> resolution from a run in Phaser, either in NCS mode or in MR_AUTO mode.
>
> In such a case, once the tNCS is properly characterised, Phaser places
> molecules in pairs (or higher multiples in more complex cases of
> higher-order tNCS).  That should deal automatically with cases like the
> ones Phoebe mentions, where you could temporarily reduce the cell by a
> factor of two, but it works more generally when the tNCS doesn't correspond
> to a cell doubling.  We'd appreciate hearing about any cases where
> reindexing in a smaller cell works and Phaser's automated treatment
> didn't!  Or, indeed, about cases like the one Ethan mentioned where turning
> off the automatic tNCS correction helped.  (I've already been in touch with
> Ethan off-line.)
>
> As Jacob and others have mentioned, you will always get lower R-factors
> once you treat the data as being twinned, and the more twin operators the
> bigger the reduction in R-factors.  So you need very strong evidence,
> independent of R-factors, to invoke twinning.  In this case, the L-test
> should be reasonably trustworthy even in the presence of tNCS, and your
> L-test values are close to what one would expect for an untwinned crystal.
> At most you have partial twinning, or perhaps twinning of a
> pseudo-symmetric crystal (a possibility Phil mentioned), where the effects
> on intensity statistics are reduced.
>
> One way to address a problem like this is to solve the structure in a
> lower symmetry space group (as you have done), but then to check whether
> the MR solution actually obeys the higher symmetry.  You can do this by
> looking at the crystal packing or at merging statistics for Fcalcs after a
> refinement with highly restrained NCS.  A similar sort of analysis is
> automated in the Zanuda tool in CCP4.
>
> Dealing with potential complications from combinations of twinning and
> pseudosymmetry is one of the more challenging aspects of crystallography,
> but it's a good learning experience.  Good luck!
>
> Randy Read
>
> > On 10 Jan 2019, at 22:38, Donghyuk Shin  wrote:
> >
> > Dear all,
> >
> > Thank you very much for all of your suggestions and sharing experiences.
> > As many of you commented, the current small unit cell C2 refinement
> seems to be incorrect or correct, and I should put some efforts to crack
> this question.
> >
> > - To Phill Jeffrey,
> > The idea, trying to find high symmetry SG with small unit cell C2 data
> is good idea, and I will try this.
> > For your last comments, identifiable electron density differences
> between each chain,
> > I guess there should not be other densities between chains if my current
> SG and model is correct. Am I right?
> >
> > - To Ethan,
> > Turning off the automatic_tNCS_option seems to be good option.
> > I think, my current data seems to be twinned then tNCS which I am not
> sure at this moment. But I will keep your advice in my mind.
> >
> > - To Phoebe A. Rice,
> > It is quite interesting that you also could get structure solution by
> indexing strong spots and having smaller unit cell.
> > Actually, I was wondering how it was possible that having half-sized
> unit cell could have solution, while full-sized unit cell could not.
> > It will be great if you can share your experience a bit more (e.g the
> size of smaller unit cell used in initial search for both 1szp and 3pkz)
> >
> > Again, thank you very much for all of your suggestion.
> >
> > Best wishes,
> > Donghyuk
> >
> > 
> >
> > To unsubscribe from the CCP4BB list, click the following link:
> > https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1
>
> --
> Randy J. Read
> Department of Haematology, University of Cambridge
> Cambridge Institute for Medical Research  Tel: + 44 1223 336500
> Wellcome Trust/MRC Building   Fax: + 44 1223 336827
> Hills Road

Re: [ccp4bb] translational NCS & twinning

[Randy Read]

Hi,

I'm just catching up on the BB after the CCP4 Study Weekend!

Going back to the beginning of this thread, it's true that tNCS and twinning 
have opposite effects on the intensity moments, which can mask each other.  
However, for simple cases (as this one appears to be) with a single NCS 
translation, the tNCS analysis in Phaser seems to be pretty successful at 
deconvoluting those effects.  It would be interesting to see the overall second 
moments and the second moments as a function of resolution from a run in 
Phaser, either in NCS mode or in MR_AUTO mode.

In such a case, once the tNCS is properly characterised, Phaser places 
molecules in pairs (or higher multiples in more complex cases of higher-order 
tNCS).  That should deal automatically with cases like the ones Phoebe 
mentions, where you could temporarily reduce the cell by a factor of two, but 
it works more generally when the tNCS doesn't correspond to a cell doubling.  
We'd appreciate hearing about any cases where reindexing in a smaller cell 
works and Phaser's automated treatment didn't!  Or, indeed, about cases like 
the one Ethan mentioned where turning off the automatic tNCS correction helped. 
 (I've already been in touch with Ethan off-line.)

As Jacob and others have mentioned, you will always get lower R-factors once 
you treat the data as being twinned, and the more twin operators the bigger the 
reduction in R-factors.  So you need very strong evidence, independent of 
R-factors, to invoke twinning.  In this case, the L-test should be reasonably 
trustworthy even in the presence of tNCS, and your L-test values are close to 
what one would expect for an untwinned crystal.  At most you have partial 
twinning, or perhaps twinning of a pseudo-symmetric crystal (a possibility Phil 
mentioned), where the effects on intensity statistics are reduced.

One way to address a problem like this is to solve the structure in a lower 
symmetry space group (as you have done), but then to check whether the MR 
solution actually obeys the higher symmetry.  You can do this by looking at the 
crystal packing or at merging statistics for Fcalcs after a refinement with 
highly restrained NCS.  A similar sort of analysis is automated in the Zanuda 
tool in CCP4.

Dealing with potential complications from combinations of twinning and 
pseudosymmetry is one of the more challenging aspects of crystallography, but 
it's a good learning experience.  Good luck!

Randy Read

> On 10 Jan 2019, at 22:38, Donghyuk Shin  wrote:
> 
> Dear all,
> 
> Thank you very much for all of your suggestions and sharing experiences.
> As many of you commented, the current small unit cell C2 refinement seems to 
> be incorrect or correct, and I should put some efforts to crack this question.
> 
> - To Phill Jeffrey, 
> The idea, trying to find high symmetry SG with small unit cell C2 data is 
> good idea, and I will try this.
> For your last comments, identifiable electron density differences between 
> each chain, 
> I guess there should not be other densities between chains if my current SG 
> and model is correct. Am I right?
> 
> - To Ethan,
> Turning off the automatic_tNCS_option seems to be good option.
> I think, my current data seems to be twinned then tNCS which I am not sure at 
> this moment. But I will keep your advice in my mind.
> 
> - To Phoebe A. Rice,
> It is quite interesting that you also could get structure solution by 
> indexing strong spots and having smaller unit cell.
> Actually, I was wondering how it was possible that having half-sized unit 
> cell could have solution, while full-sized unit cell could not.
> It will be great if you can share your experience a bit more (e.g the size of 
> smaller unit cell used in initial search for both 1szp and 3pkz)
> 
> Again, thank you very much for all of your suggestion.
> 
> Best wishes,
> Donghyuk
> 
> 
> 
> To unsubscribe from the CCP4BB list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1

--
Randy J. Read
Department of Haematology, University of Cambridge
Cambridge Institute for Medical Research  Tel: + 44 1223 336500
Wellcome Trust/MRC Building   Fax: + 44 1223 336827
Hills RoadE-mail: rj...@cam.ac.uk
Cambridge CB2 0XY, U.K.   www-structmed.cimr.cam.ac.uk



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1

Re: [ccp4bb] AW: [ccp4bb] AW: [EXTERNAL] [ccp4bb] translational NCS & twinning

[IGBMC]

Le Vendredi 11 Janvier 2019 09:07 CET, herman.schreu...@sanofi.com a écrit:

Dear Herman,
As far as I understood the twinning problem, what you say is only true in some 
occasions, and not in others.
If the "macroscopic" domains are so small that they are smaller than the X-rays 
coherence length, then you may do what you say because the X-rays emitted by 
the different domains can interfere.
But, if the domains become large, the X-rays emitted by the different domains 
do not interfere anymore and you have to add the weighted intensities, not the 
amplitudes, of each domain.
I hope I understood well your comment and did not "interfere negatively" with 
the thread...
Best
Philippe Dumas

> Dear Lan,
>
> Thank you for your compliment. I do not use Xtriage, so I did not bother 
> looking at the log files.
>
> What I meant to say is that with twinning, the crystal has different 
> macroscopic domains where the molecules have different orientations, say one 
> domain with orientation A and one domain with orientation B. Since these 
> domains grow on top of each other, they are usually related by a twin 
> operator similar to a crystallographic operator such as a twofold axis.
> The fourier transform of the electron density of the crystal is the 
> convolution of the fourier transform of the individual molecules with the 
> crystal lattice, with the fourier transform of the individual molecules 
> usually giving the stronger contribution. So to get a solution with a decent 
> R-factor, one must include all orientations (A, B etc.) in the model, with 
> the position of the molecules in the crystal lattice contributing less to the 
> diffraction pattern. So one can put the orientations on top of each other in 
> a small unit cell using twinning, or put them in a larger unit cell at 
> different positions using crystallographic or non-crystallographic symmetry. 
> That is what I meant be "twinning" (N)CS.
>
> Hope this makes my remark a little clearer.
>
> Best,
> Herman
>
> PS: While other BB readers may have had the same question, I have posted the 
> reply to the BB. I hope you don't mind.
>
>
> -Ursprüngliche Nachricht-
> Von: Guan, Lan [mailto:lan.g...@ttuhsc.edu]
> Gesendet: Donnerstag, 10. Januar 2019 20:53
> An: Schreuder, Herman /DE
> Betreff: Re: [ccp4bb] AW: [EXTERNAL] [ccp4bb] translational NCS & twinning
>
> Dear Herman,
> I have read your insightful comments on twining and tNCS for years now, which 
> is very useful and helpful!  Thanks,
>
> For Donghyuk’s case, do you think that he really has a twinning issue?  With 
> a lower symmetry, all possible twining operators is always reported in the 
> Xtriage and did not mean a real twin existed.  His L test shows a twin 
> fraction of 0.00 in his log file.  The intensity statistics does not really 
> indicate an actually twinning.  Base on the refinement, twin law is needed to 
> get refinement going.  It looks like a twin.  I am confused...
>
> > the molecules are related by "twinning" (N)CS?
>
>
> What does this mean “ twinning (N)CS"?  Would you please kindly explain it 
> further?
>
> Thanks,
>
>
> Lan
>
>
> 
> Lan Guan, MD PhD
> Associate Professor | Department of Cell Physiology and Molecular Biophysics
> Director | Center for Membrane Protein Research
>
> 3601 4th St. MS 6551 | Lubbock, TX 79430
> 5A148A (Office) | (1) 806 743-3102 (Phone) | lan.g...@ttuhsc.edu (E-Mail)
>
> http://www.ttuhsc.edu/medicine/cell-physiology-molecular-biophysics/faculty/guan/
> https://www.ttuhsc.edu/centers-institutes/membrane-protein-research/

> 
>
> > On Jan 10, 2019, at 11:10 AM, herman.schreu...@sanofi.com wrote:
> >
> > CAUTION: This email originated from outside of TTUHSC. Do not click links 
> > or open attachments unless you recognize the sender and know the content is 
> > safe.
> >
> >
> > Dear Donghyuk,
> >
> > Unfortunately, everything is possible when NCS, twinning etc. get into the 
> > game. I do not have answers, but some questions for you to think about:
> > - Do you really have 6 twinning operators, or only one operator and are the 
> > other operators generated by (non)crystallographic symmetry?
> > - Both your P6322 cell and your C2 cell have angles of 90 90 90. For P6322, 
> > the last angle should be 120 and for C2, only the second angle is 
> > constrained to be 90. Maybe you should check that not somewhere something 
> > went wrong with the cell angles.
> > - How weak are the reflecti

[ccp4bb] AW: [ccp4bb] AW: [EXTERNAL] [ccp4bb] translational NCS & twinning

[Herman . Schreuder]

Dear Lan,

Thank you for your compliment. I do not use Xtriage, so I did not bother 
looking at the log files.

What I meant to say is that with twinning, the crystal has different 
macroscopic domains where the molecules have different orientations, say one 
domain with orientation A and one domain with orientation B. Since these 
domains grow on top of each other, they are usually related by a twin operator 
similar to a crystallographic operator such as a twofold axis. 
The fourier transform of the electron density of the crystal is the convolution 
of the fourier transform of the individual molecules with the crystal lattice, 
with the fourier transform of the individual molecules usually giving the 
stronger contribution. So to get a solution with a decent R-factor, one must 
include all orientations (A, B etc.) in the model, with the position of the 
molecules in the crystal lattice contributing less to the diffraction pattern. 
So one can put the orientations on top of each other in a small unit cell using 
twinning, or put them in a larger unit cell at different positions using 
crystallographic or non-crystallographic symmetry. That is what I meant be 
"twinning" (N)CS.

Hope this makes my remark a little clearer.

Best,
Herman

PS: While other BB readers may have had the same question, I have posted the 
reply to the BB. I hope you don't mind.


-Ursprüngliche Nachricht-
Von: Guan, Lan [mailto:lan.g...@ttuhsc.edu] 
Gesendet: Donnerstag, 10. Januar 2019 20:53
An: Schreuder, Herman /DE
Betreff: Re: [ccp4bb] AW: [EXTERNAL] [ccp4bb] translational NCS & twinning

Dear Herman,
I have read your insightful comments on twining and tNCS for years now, which 
is very useful and helpful!  Thanks,

For Donghyuk’s case, do you think that he really has a twinning issue?  With a 
lower symmetry, all possible twining operators is always reported in the 
Xtriage and did not mean a real twin existed.  His L test shows a twin fraction 
of 0.00 in his log file.  The intensity statistics does not really indicate an 
actually twinning.  Base on the refinement, twin law is needed to get 
refinement going.  It looks like a twin.  I am confused...

> the molecules are related by "twinning" (N)CS?


What does this mean “ twinning (N)CS"?  Would you please kindly explain it 
further?

Thanks,


Lan



Lan Guan, MD PhD
Associate Professor | Department of Cell Physiology and Molecular Biophysics
Director | Center for Membrane Protein Research

3601 4th St. MS 6551 | Lubbock, TX 79430
5A148A (Office) | (1) 806 743-3102 (Phone) | lan.g...@ttuhsc.edu (E-Mail)

http://www.ttuhsc.edu/medicine/cell-physiology-molecular-biophysics/faculty/guan/
https://www.ttuhsc.edu/centers-institutes/membrane-protein-research/


> On Jan 10, 2019, at 11:10 AM, herman.schreu...@sanofi.com wrote:
> 
> CAUTION: This email originated from outside of TTUHSC. Do not click links or 
> open attachments unless you recognize the sender and know the content is safe.
> 
> 
> Dear Donghyuk,
> 
> Unfortunately, everything is possible when NCS, twinning etc. get into the 
> game. I do not have answers, but some questions for you to think about:
> - Do you really have 6 twinning operators, or only one operator and are the 
> other operators generated by (non)crystallographic symmetry?
> - Both your P6322 cell and your C2 cell have angles of 90 90 90. For P6322, 
> the last angle should be 120 and for C2, only the second angle is constrained 
> to be 90. Maybe you should check that not somewhere something went wrong with 
> the cell angles.
> - How weak are the reflections that got discarded by halving the a- and 
> b-axes? Do they have significant intensity, or is it only noise?
> - By shrinking the unit cell, you may have created artificial twinning when 
> in the large unit cell the molecules are related by "twinning" (N)CS.
> - Since you seem to have found a solution with the small unit cell, you could 
> see if you could fit this solution in the large unit cell: Process in P1 in 
> the large unit cell and use the ensemble (the complete! unit cell of your C2 
> solution, as a search model.
> - Your current solution maybe correct after all, but I would analyze it very 
> critically.
> 
> Best,
> Herman
> 
> 
> -Ursprüngliche Nachricht-
> Von: CCP4 bulletin board [mailto:CCP4BB@JISCMAIL.AC.UK] Im Auftrag von 
> Donghyuk Shin
> Gesendet: Donnerstag, 10. Januar 2019 11:12
> An: CCP4BB@JISCMAIL.AC.UK
> Betreff: [EXTERNAL] [ccp4bb] translational NCS & twinning
> 
> Dear all,
> 
> I am having tough time with my Xtal data sets those seem to be twinned or 
> have translational NCS, and it will be grea

Re: [ccp4bb] translational NCS & twinning

[Donghyuk Shin]

Dear all,

Thank you very much for all of your suggestions and sharing experiences.
As many of you commented, the current small unit cell C2 refinement seems to be 
incorrect or correct, and I should put some efforts to crack this question.

- To Phill Jeffrey, 
The idea, trying to find high symmetry SG with small unit cell C2 data is good 
idea, and I will try this.
For your last comments, identifiable electron density differences between each 
chain, 
I guess there should not be other densities between chains if my current SG and 
model is correct. Am I right?

- To Ethan,
Turning off the automatic_tNCS_option seems to be good option.
I think, my current data seems to be twinned then tNCS which I am not sure at 
this moment. But I will keep your advice in my mind.

- To Phoebe A. Rice,
It is quite interesting that you also could get structure solution by indexing 
strong spots and having smaller unit cell.
Actually, I was wondering how it was possible that having half-sized unit cell 
could have solution, while full-sized unit cell could not.
It will be great if you can share your experience a bit more (e.g the size of 
smaller unit cell used in initial search for both 1szp and 3pkz)

Again, thank you very much for all of your suggestion.

Best wishes,
Donghyuk



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1

Re: [ccp4bb] translational NCS & twinning

[Phoebe A. Rice]

For two projects in the distant past, we dealt with tNCS by initially telling 
lies to the software (1szp and 3pkz).  The tNCS was strong enough that there 
was a clear weak/dark pattern in the diffraction pattern, so for the initial 
molecular replacement we used a data set in the smaller unit cell where only 
the strong spots had been boxed and reduced.  Then we moved that model to the 
proper, full unit cell (weak spots included) data set and finished the 
molecular replacement either by manually pushing the model around according to 
the appropriate vector(s) or just doing more molrep.

If you're desperate, and your weak spots are weak enough, it is worth a try!

~~~
Phoebe A. Rice
Dept. of Biochem & Mol. Biol. and
  Committee on Microbiology
https://voices.uchicago.edu/phoebericelab/
 
 

On 1/10/19, 12:18 PM, "CCP4 bulletin board on behalf of Ethan A Merritt" 
 wrote:

On Thursday, January 10, 2019 2:11:52 AM PST Donghyuk Shin wrote:
> Dear all,
> 
> I am having tough time with my Xtal data sets those seem to be twinned or 
have translational NCS, and it will be greatly appreciated if you can give me 
some advices or comments!
> 
> Data was initially processed with XDS and scaled with aimless without 
specifying certain space group (SG).
> Aimless picked the P 63 2 2 for the best SG, but the xtriage indicates 
there is non-origin peak after patterson analysis. (attached log)
> And, I could not get the proper MR-solution from this data sets.
> 
> Because I read that twinning and tNCS cannot be properly distinguished at 
high SG, I went down to subgroup either P32 or P6 assuming that there is 
twinning which make data set seems to have apparently high SG. (procedure was 
same XDS->aimless but I specified the SG to keep them)
> Then, xtriage still indicates there is non-origin peak as before, but 
found twin laws for the data sets (attached log).
> However, I still could not get the right MR-solution.
> Then, I went even further down to P3 or C2, and xtriage found more twin 
laws which is make sense because of the lower SG. (attached log)

> Again, I could not get the MR-solution.
> For all the MR running above, I assumed that phaser(ccp4 module) 
automatically applied tNCS if they present. or should I have to tick on button 
in the expert parameters?
   ^^^

Hi Donghuk,

This may indeed be a possible source of problems.   I have recent 
experience with a
case where aimless/xtriage/phaser all report a large tNCS component, but 
allowing
phaser to use this information prevents finding a correct MR solution.  I 
can only
obtain the known correct solution by telling phaser explicitly _not_ to use 
tNCS.
In my case the true space group is P1 but the angles are all close to 90., 
causing
most indexing programs to falsely identify it as P2.  I cannot say how or 
why the
tNCS test triggers, but the known correct solution in P1 does not contain 
tNCS.

Ethan


> 
> Then, I went back to the image and processed the datasets with mosflm by 
checking the indexed spots.
> During this step, I played with the threshold for indexing to follow the 
strong spot for get correct SG.
> I am not sure whether this is correct or not, but by putting high 
threshold for indexing (e.g. ~15) I could index the data with C2 which has half 
dimension for a,b axes (116.348,  67.218, 182.861,  90, 90, 90) to the original 
unit cell (232.533, 134.202, 182.67, 90, 90, 90).
> With this, I could put 3 molecules in ASU by MR. During refinement, I 
felt that the R values were not dropping, and I applied twin refinement.
> without twin refinement the R values were (0.39/0.44, work/free), and 
applying twin refinement gave me significantly better values (0.23/0.26).
> Because there were 6 twin operators which may cause this huge R value 
drop, I speculate whether this is true or not.
> 
> Your comments will be greatly helpful! 
> 
> With you all the best,
> Donghyuk
> 
> 
> 
> 
> 
> To unsubscribe from the CCP4BB list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1


-- 
Ethan A Merritt
Biomolecular Structure Center,  K-428 Health Sciences Bldg
MS 357742,   University of Washington, Seattle 98195-7742



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1





To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1

Re: [ccp4bb] translational NCS & twinning

[Ethan A Merritt]

On Thursday, January 10, 2019 2:11:52 AM PST Donghyuk Shin wrote:
> Dear all,
> 
> I am having tough time with my Xtal data sets those seem to be twinned or 
> have translational NCS, and it will be greatly appreciated if you can give me 
> some advices or comments!
> 
> Data was initially processed with XDS and scaled with aimless without 
> specifying certain space group (SG).
> Aimless picked the P 63 2 2 for the best SG, but the xtriage indicates there 
> is non-origin peak after patterson analysis. (attached log)
> And, I could not get the proper MR-solution from this data sets.
> 
> Because I read that twinning and tNCS cannot be properly distinguished at 
> high SG, I went down to subgroup either P32 or P6 assuming that there is 
> twinning which make data set seems to have apparently high SG. (procedure was 
> same XDS->aimless but I specified the SG to keep them)
> Then, xtriage still indicates there is non-origin peak as before, but found 
> twin laws for the data sets (attached log).
> However, I still could not get the right MR-solution.
> Then, I went even further down to P3 or C2, and xtriage found more twin laws 
> which is make sense because of the lower SG. (attached log)

> Again, I could not get the MR-solution.
> For all the MR running above, I assumed that phaser(ccp4 module) 
> automatically applied tNCS if they present. or should I have to tick on 
> button in the expert parameters?
   ^^^

Hi Donghuk,

This may indeed be a possible source of problems.   I have recent experience 
with a
case where aimless/xtriage/phaser all report a large tNCS component, but 
allowing
phaser to use this information prevents finding a correct MR solution.  I can 
only
obtain the known correct solution by telling phaser explicitly _not_ to use 
tNCS.
In my case the true space group is P1 but the angles are all close to 90., 
causing
most indexing programs to falsely identify it as P2.  I cannot say how or why 
the
tNCS test triggers, but the known correct solution in P1 does not contain tNCS.

Ethan


> 
> Then, I went back to the image and processed the datasets with mosflm by 
> checking the indexed spots.
> During this step, I played with the threshold for indexing to follow the 
> strong spot for get correct SG.
> I am not sure whether this is correct or not, but by putting high threshold 
> for indexing (e.g. ~15) I could index the data with C2 which has half 
> dimension for a,b axes (116.348,  67.218, 182.861,  90, 90, 90) to the 
> original unit cell (232.533, 134.202, 182.67, 90, 90, 90).
> With this, I could put 3 molecules in ASU by MR. During refinement, I felt 
> that the R values were not dropping, and I applied twin refinement.
> without twin refinement the R values were (0.39/0.44, work/free), and 
> applying twin refinement gave me significantly better values (0.23/0.26).
> Because there were 6 twin operators which may cause this huge R value drop, I 
> speculate whether this is true or not.
> 
> Your comments will be greatly helpful! 
> 
> With you all the best,
> Donghyuk
> 
> 
> 
> 
> 
> To unsubscribe from the CCP4BB list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1


-- 
Ethan A Merritt
Biomolecular Structure Center,  K-428 Health Sciences Bldg
MS 357742,   University of Washington, Seattle 98195-7742



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1

Re: [ccp4bb] translational NCS & twinning

[Phil Jeffrey]

Donghyuk

The combination of two things gives me cause for concern:
1.  You've reindexed something that apparently scaled OK in point group 
622 into point group 2, with a smaller cell.  Since it's hard to fake 
that sort of data agreement in 622, I assume your data is at the very 
least pseudo-622.
2.  You've modeled that additional symmetry using a whole array of twin 
operators and some non-crystallographic symmetry.


This may in fact be the correct model, but there's a significant risk 
that you're inappropriately modeling something.


Let's assume for a moment that your small-cell C2 refinement with 6 twin 
ops improves the model somewhat.  Now go back and generate scaled data 
sets in all possible point groups suggested by Pointless for that data, 
and try and find molecular replacement solutions with your 
partially-refined model by testing every possible space group in every 
possible point group based on the Pointless suggestions.  The idea is to 
try and model more of the 622 pseudo-symmetry as crystallographic 
symmetry, using fewer twin operators in refinement.  If you test all of 
these combinations, which might be quite extensive, you might find one 
or more that fit nearly as well as your current C2 solution and has 
higher symmetry. You should take a hard look at that those potential 
solutions.  Only when you've thoroughly exhausted alternative molecular 
replacement solutions would you be confident that your C2 model is in 
fact the only reasonable explanation of your data.


But as it stands it is rather atypical and it warrants further investigation

Additional evidence that your C2 cell is the only reasonable model would 
be identifiable electron density differences between each chain in your 
(presumed) multi-chain model.


Cheers
Phil Jeffrey
Princeton

On 1/10/19 11:08 AM, Donghyuk Shin wrote:

Dear Jacob Keller and Vipul,

Thank you both very much for the reply.
Regarding the R-values, I am just wondering whether the huge gap between 
refinements w/ w/o twin operator can be possible even the crystal is not 
twinned?

Best wishes,
Donghyuk



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1





To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1

[ccp4bb] AW: [EXTERNAL] [ccp4bb] translational NCS & twinning

[Herman . Schreuder]

Dear Donghyuk,

Unfortunately, everything is possible when NCS, twinning etc. get into the 
game. I do not have answers, but some questions for you to think about:
- Do you really have 6 twinning operators, or only one operator and are the 
other operators generated by (non)crystallographic symmetry?
- Both your P6322 cell and your C2 cell have angles of 90 90 90. For P6322, the 
last angle should be 120 and for C2, only the second angle is constrained to be 
90. Maybe you should check that not somewhere something went wrong with the 
cell angles.
- How weak are the reflections that got discarded by halving the a- and b-axes? 
Do they have significant intensity, or is it only noise?
- By shrinking the unit cell, you may have created artificial twinning when in 
the large unit cell the molecules are related by "twinning" (N)CS.
- Since you seem to have found a solution with the small unit cell, you could 
see if you could fit this solution in the large unit cell: Process in P1 in the 
large unit cell and use the ensemble (the complete! unit cell of your C2 
solution, as a search model.
- Your current solution maybe correct after all, but I would analyze it very 
critically.

Best,
Herman


-Ursprüngliche Nachricht-
Von: CCP4 bulletin board [mailto:CCP4BB@JISCMAIL.AC.UK] Im Auftrag von Donghyuk 
Shin
Gesendet: Donnerstag, 10. Januar 2019 11:12
An: CCP4BB@JISCMAIL.AC.UK
Betreff: [EXTERNAL] [ccp4bb] translational NCS & twinning

Dear all,

I am having tough time with my Xtal data sets those seem to be twinned or have 
translational NCS, and it will be greatly appreciated if you can give me some 
advices or comments!

Data was initially processed with XDS and scaled with aimless without 
specifying certain space group (SG).
Aimless picked the P 63 2 2 for the best SG, but the xtriage indicates there is 
non-origin peak after patterson analysis. (attached log) And, I could not get 
the proper MR-solution from this data sets.

Because I read that twinning and tNCS cannot be properly distinguished at high 
SG, I went down to subgroup either P32 or P6 assuming that there is twinning 
which make data set seems to have apparently high SG. (procedure was same 
XDS->aimless but I specified the SG to keep them) Then, xtriage still indicates 
there is non-origin peak as before, but found twin laws for the data sets 
(attached log).
However, I still could not get the right MR-solution.
Then, I went even further down to P3 or C2, and xtriage found more twin laws 
which is make sense because of the lower SG. (attached log) Again, I could not 
get the MR-solution.
For all the MR running above, I assumed that phaser(ccp4 module) automatically 
applied tNCS if they present. or should I have to tick on button in the expert 
parameters?

Then, I went back to the image and processed the datasets with mosflm by 
checking the indexed spots.
During this step, I played with the threshold for indexing to follow the strong 
spot for get correct SG.
I am not sure whether this is correct or not, but by putting high threshold for 
indexing (e.g. ~15) I could index the data with C2 which has half dimension for 
a,b axes (116.348,  67.218, 182.861,  90, 90, 90) to the original unit cell 
(232.533, 134.202, 182.67, 90, 90, 90).
With this, I could put 3 molecules in ASU by MR. During refinement, I felt that 
the R values were not dropping, and I applied twin refinement.
without twin refinement the R values were (0.39/0.44, work/free), and applying 
twin refinement gave me significantly better values (0.23/0.26).
Because there were 6 twin operators which may cause this huge R value drop, I 
speculate whether this is true or not.

Your comments will be greatly helpful! 

With you all the best,
Donghyuk





To unsubscribe from the CCP4BB list, click the following link:
https://urldefense.proofpoint.com/v2/url?u=https-3A__www.jiscmail.ac.uk_cgi-2Dbin_webadmin-3FSUBED1-3DCCP4BB-26A-3D1=DwIFaQ=Dbf9zoswcQ-CRvvI7VX5j3HvibIuT3ZiarcKl5qtMPo=HK-CY_tL8CLLA93vdywyu3qI70R4H8oHzZyRHMQu1AQ=xlZKtKm2eNS0qmQb4wXnIIknGL1-n0M0toJoWb2Ezco=scEheHieLBLhJMaXoEDimlYM9TAk600XwBjYXrQC1nc=



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1

Re: [ccp4bb] translational NCS & twinning

[Donghyuk Shin]

Dear Jacob Keller and Vipul,

Thank you both very much for the reply.
Regarding the R-values, I am just wondering whether the huge gap between 
refinements w/ w/o twin operator can be possible even the crystal is not 
twinned?

Best wishes,
Donghyuk



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1

Re: [ccp4bb] translational NCS & twinning

[Keller, Jacob]

>>I feel you went ahead with right strategy.

I agree with this part regarding lowering symmetry.

>>For 2.1 A datasrt, the appropriate drop in Rfree/ Rwork is a strong 
>>indicator, i believe.

This is not true—even non-twinned data will improve in R values with twinning 
operators added as parameters. And twin-refined structures always have lower R 
values.
JPK


On Thu 10 Jan, 2019, 3:52 PM Donghyuk Shin 
mailto:sdh...@gmail.com> wrote:
Dear all,

I am having tough time with my Xtal data sets those seem to be twinned or have 
translational NCS, and it will be greatly appreciated if you can give me some 
advices or comments!

Data was initially processed with XDS and scaled with aimless without 
specifying certain space group (SG).
Aimless picked the P 63 2 2 for the best SG, but the xtriage indicates there is 
non-origin peak after patterson analysis. (attached log)
And, I could not get the proper MR-solution from this data sets.

Because I read that twinning and tNCS cannot be properly distinguished at high 
SG, I went down to subgroup either P32 or P6 assuming that there is twinning 
which make data set seems to have apparently high SG. (procedure was same 
XDS->aimless but I specified the SG to keep them)
Then, xtriage still indicates there is non-origin peak as before, but found 
twin laws for the data sets (attached log).
However, I still could not get the right MR-solution.
Then, I went even further down to P3 or C2, and xtriage found more twin laws 
which is make sense because of the lower SG. (attached log)
Again, I could not get the MR-solution.
For all the MR running above, I assumed that phaser(ccp4 module) automatically 
applied tNCS if they present. or should I have to tick on button in the expert 
parameters?

Then, I went back to the image and processed the datasets with mosflm by 
checking the indexed spots.
During this step, I played with the threshold for indexing to follow the strong 
spot for get correct SG.
I am not sure whether this is correct or not, but by putting high threshold for 
indexing (e.g. ~15) I could index the data with C2 which has half dimension for 
a,b axes (116.348,  67.218, 182.861,  90, 90, 90) to the original unit cell 
(232.533, 134.202, 182.67, 90, 90, 90).
With this, I could put 3 molecules in ASU by MR. During refinement, I felt that 
the R values were not dropping, and I applied twin refinement.
without twin refinement the R values were (0.39/0.44, work/free), and applying 
twin refinement gave me significantly better values (0.23/0.26).
Because there were 6 twin operators which may cause this huge R value drop, I 
speculate whether this is true or not.

Your comments will be greatly helpful!

With you all the best,
Donghyuk





To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1

Re: [ccp4bb] translational NCS & twinning

[vipul panchal]

Hi Donghyuk,
I feel you went ahead with right strategy.
For 2.1 A datasrt, the appropriate drop in Rfree/ Rwork is a strong
indicator, i believe.
If you have already build all possible model, using tls can be of further
help.

Cheers,
Vipul




On Thu 10 Jan, 2019, 3:52 PM Donghyuk Shin  Dear all,
>
> I am having tough time with my Xtal data sets those seem to be twinned or
> have translational NCS, and it will be greatly appreciated if you can give
> me some advices or comments!
>
> Data was initially processed with XDS and scaled with aimless without
> specifying certain space group (SG).
> Aimless picked the P 63 2 2 for the best SG, but the xtriage indicates
> there is non-origin peak after patterson analysis. (attached log)
> And, I could not get the proper MR-solution from this data sets.
>
> Because I read that twinning and tNCS cannot be properly distinguished at
> high SG, I went down to subgroup either P32 or P6 assuming that there is
> twinning which make data set seems to have apparently high SG. (procedure
> was same XDS->aimless but I specified the SG to keep them)
> Then, xtriage still indicates there is non-origin peak as before, but
> found twin laws for the data sets (attached log).
> However, I still could not get the right MR-solution.
> Then, I went even further down to P3 or C2, and xtriage found more twin
> laws which is make sense because of the lower SG. (attached log)
> Again, I could not get the MR-solution.
> For all the MR running above, I assumed that phaser(ccp4 module)
> automatically applied tNCS if they present. or should I have to tick on
> button in the expert parameters?
>
> Then, I went back to the image and processed the datasets with mosflm by
> checking the indexed spots.
> During this step, I played with the threshold for indexing to follow the
> strong spot for get correct SG.
> I am not sure whether this is correct or not, but by putting high
> threshold for indexing (e.g. ~15) I could index the data with C2 which has
> half dimension for a,b axes (116.348,  67.218, 182.861,  90, 90, 90) to the
> original unit cell (232.533, 134.202, 182.67, 90, 90, 90).
> With this, I could put 3 molecules in ASU by MR. During refinement, I felt
> that the R values were not dropping, and I applied twin refinement.
> without twin refinement the R values were (0.39/0.44, work/free), and
> applying twin refinement gave me significantly better values (0.23/0.26).
> Because there were 6 twin operators which may cause this huge R value
> drop, I speculate whether this is true or not.
>
> Your comments will be greatly helpful!
>
> With you all the best,
> Donghyuk
>
>
>
> 
>
> To unsubscribe from the CCP4BB list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1
>



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1