Re: [ccp4bb] AW: [EXTERNAL] Re: [ccp4bb] Lattice-translocation defect (LTD)

2020-02-12 Thread Eleanor Dodson 
I am no expert, but a) a very strong peak 7A from the origin means two
molecules 7A apart?? Most unlikely ..

The first thing to look at is the actual images - Lattice translation
defects usually generate very streaky patterns.
Integration programs can cleverly select a lattice and ignore the
unpredicted pixels - (by the way - why don't the data integration programs
scream when this occurs? )
LTDs usually produce some strange intensity statistics too - what do the
moments/twin tests etc look like?


Otherwise your spacegroup must be wrong, although providing you have got
the right crystalclass, I cant think how that can generate such a Patterson
peak..

Good luck Eleanor
PS - maybe search out another crystal!

On Wed, 12 Feb 2020 at 10:25, Harry Powell - CCP4BB <
193323b1e616-dmarc-requ...@jiscmail.ac.uk> wrote:

> Hi
>
> Something else I should have mentioned - in iMosflm you can sum your
> images for viewing only if you have them as  HDF5 or Pilatus CBF (as well
> as summing them for processing if you have HDF5).
>
> Harry
>
> On 12 Feb 2020, at 10:18, Schreuder, Herman /DE <
> herman.schreu...@sanofi.com> wrote:
>
> Hi Daniele,
>
> I agree with Wim that the first thing you should check is your space group
> and especially whether a ncs symmetry element has been mistakenly
> identified as being crystallographic. Since your Patterson peak is along w
> (c-axis), you have to change the space group for processing such, that
> there are no rotation axes or other symmetry elements perpendicular to the
> c-axis any more. If you have a low-symmetry space group, you could also
> process in P1 to be absolutely on the safe side. Than you should run MR in
> this lower symmetry space group.
>
> Concerning lattice translocation defects, almost every case is unique and
> I am not aware of any software able to handle this. Here you will have to
> work out the maths for your particular case yourself and apply the
> correction with e.g. sftools. I am a little puzzled by your 7Å vector in
> your (native?) Patterson maps. I guess, if you translate your protein by 7Å
> it will strongly overlap with itself and I guess the same will be true for
> your ghost map. You should also have a look at your diffraction images,
> perhaps after summing them to 1° slices so become human-interpretable. In
> many cases, LTD is associated with a mix of sharp and fuzzy diffraction
> spots. Seeing those would be a strong indicator that you have the problem,
> but there are cases of LTD where all the spots are sharp. You may also want
> to check for statistical disorder.
>
> Good luck!
> Herman
>
> *Von:* CCP4 bulletin board  *Im Auftrag von *Wim
> Burmeister
> *Gesendet:* Mittwoch, 12. Februar 2020 08:57
> *An:* CCP4BB@JISCMAIL.AC.UK
> *Betreff:* [EXTERNAL] Re: [ccp4bb] Lattice-translocation defect (LTD)
>
>
> *EXTERNAL : *Real sender is owner-ccp...@jiscmail.ac.uk
>
>
>
> Hello,
>
> do you have some details about the space group ? Did the integration not
> miss any sports ? I would rather think of an ncs close to crystallographic
> symmetry, or maybe some twinning problem.
>
> I guess these are Pilatus data, can you combine the frames into 1 degree
> oscillations and try Mosflm processing to see how the patterns integrate ?
>
> Greetings
>
> Wim
> Le 11/02/2020 à 22:31, Daniele de Sanctis a écrit :
>
> Hi all,
>
> I'm currently dealing with what I think it is a case of LTD (off-origin
> Patterson peak, with vector along w of ~ 7A and electron density map
> showing a "ghost" map shifted by 7 A). I saw there are quite a few cases
> reported in literature  (for example Hare et al, 2006), but what I could
> not find is how I can demodulate the data. Is there any software that can
> be used for this?
>
> Thank you
>
> Daniele
>
> --
> ἀρετή
> ---
> Daniele de Sanctis, PhD
>
> Structural Biology Group
> ESRF, Grenoble, France
> Tel 33 (0)4 76 88 2869
>
> --
>
> To unsubscribe from the CCP4BB list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1
> 
>
> --
> Wim Burmeister
>
> Professeur
> Institut de Biologie Structurale (IBS) CIBB
> 71 avenue des Martyrs / CS 20192
> 38044 Grenoble Cedex 9, FRANCE
> E-mail: wim.burmeis...@ibs.fr
> Tel:+33 (0) 457 42 87 41   Fax: +33 (0) 476 20 94 00
> website
>

Re: [ccp4bb] AW: [EXTERNAL] Re: [ccp4bb] Lattice-translocation defect (LTD)

2020-02-12 Thread Harry Powell - CCP4BB 
Hi

Something else I should have mentioned - in iMosflm you can sum your images for 
viewing only if you have them as  HDF5 or Pilatus CBF (as well as summing them 
for processing if you have HDF5).

Harry

> On 12 Feb 2020, at 10:18, Schreuder, Herman /DE  
> wrote:
> 
> Hi Daniele,
>  
> I agree with Wim that the first thing you should check is your space group 
> and especially whether a ncs symmetry element has been mistakenly identified 
> as being crystallographic. Since your Patterson peak is along w (c-axis), you 
> have to change the space group for processing such, that there are no 
> rotation axes or other symmetry elements perpendicular to the c-axis any 
> more. If you have a low-symmetry space group, you could also process in P1 to 
> be absolutely on the safe side. Than you should run MR in this lower symmetry 
> space group.
>  
> Concerning lattice translocation defects, almost every case is unique and I 
> am not aware of any software able to handle this. Here you will have to work 
> out the maths for your particular case yourself and apply the correction with 
> e.g. sftools. I am a little puzzled by your 7Å vector in your (native?) 
> Patterson maps. I guess, if you translate your protein by 7Å it will strongly 
> overlap with itself and I guess the same will be true for your ghost map. You 
> should also have a look at your diffraction images, perhaps after summing 
> them to 1° slices so become human-interpretable. In many cases, LTD is 
> associated with a mix of sharp and fuzzy diffraction spots. Seeing those 
> would be a strong indicator that you have the problem, but there are cases of 
> LTD where all the spots are sharp. You may also want to check for statistical 
> disorder.
>  
> Good luck!
> Herman
>  
> Von: CCP4 bulletin board  > Im Auftrag von Wim Burmeister
> Gesendet: Mittwoch, 12. Februar 2020 08:57
> An: CCP4BB@JISCMAIL.AC.UK 
> Betreff: [EXTERNAL] Re: [ccp4bb] Lattice-translocation defect (LTD)
>  
> EXTERNAL : Real sender is owner-ccp...@jiscmail.ac.uk 
> 
>  
> 
> Hello,
> 
> do you have some details about the space group ? Did the integration not miss 
> any sports ? I would rather think of an ncs close to crystallographic 
> symmetry, or maybe some twinning problem.
> 
> I guess these are Pilatus data, can you combine the frames into 1 degree 
> oscillations and try Mosflm processing to see how the patterns integrate ?
> 
> Greetings
> 
> Wim
> 
> Le 11/02/2020 à 22:31, Daniele de Sanctis a écrit :
> Hi all,
>  
> I'm currently dealing with what I think it is a case of LTD (off-origin 
> Patterson peak, with vector along w of ~ 7A and electron density map showing 
> a "ghost" map shifted by 7 A). I saw there are quite a few cases reported in 
> literature  (for example Hare et al, 2006), but what I could not find is how 
> I can demodulate the data. Is there any software that can be used for this?
>  
> Thank you
>  
> Daniele
> 
> -- 
> ἀρετή
> ---
> Daniele de Sanctis, PhD
> 
> Structural Biology Group
> ESRF, Grenoble, France
> Tel 33 (0)4 76 88 2869
>  
> To unsubscribe from the CCP4BB list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1 
> 
> -- 
> Wim Burmeister
> Professeur
> Institut de Biologie Structurale (IBS) CIBB
> 71 avenue des Martyrs / CS 20192
> 38044 Grenoble Cedex 9, FRANCE
> E-mail: wim.burmeis...@ibs.fr 
> Tel:+33 (0) 457 42 87 41   Fax: +33 (0) 476 20 94 00
> website 
> 
> map 
> 
> 
> Changement climatique : «Les autres combats n’ont aucun sens si celui-là est 
> perdu» (Aurélien Barrau)   
>  
> 
>  
> To unsubscribe from the CCP4BB list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1 
>

[ccp4bb] AW: [EXTERNAL] Re: [ccp4bb] Lattice-translocation defect (LTD)

2020-02-12 Thread Schreuder, Herman /DE 
Hi Daniele,

I agree with Wim that the first thing you should check is your space group and 
especially whether a ncs symmetry element has been mistakenly identified as 
being crystallographic. Since your Patterson peak is along w (c-axis), you have 
to change the space group for processing such, that there are no rotation axes 
or other symmetry elements perpendicular to the c-axis any more. If you have a 
low-symmetry space group, you could also process in P1 to be absolutely on the 
safe side. Than you should run MR in this lower symmetry space group.

Concerning lattice translocation defects, almost every case is unique and I am 
not aware of any software able to handle this. Here you will have to work out 
the maths for your particular case yourself and apply the correction with e.g. 
sftools. I am a little puzzled by your 7Å vector in your (native?) Patterson 
maps. I guess, if you translate your protein by 7Å it will strongly overlap 
with itself and I guess the same will be true for your ghost map. You should 
also have a look at your diffraction images, perhaps after summing them to 1° 
slices so become human-interpretable. In many cases, LTD is associated with a 
mix of sharp and fuzzy diffraction spots. Seeing those would be a strong 
indicator that you have the problem, but there are cases of LTD where all the 
spots are sharp. You may also want to check for statistical disorder.

Good luck!
Herman

Von: CCP4 bulletin board  Im Auftrag von Wim Burmeister
Gesendet: Mittwoch, 12. Februar 2020 08:57
An: CCP4BB@JISCMAIL.AC.UK
Betreff: [EXTERNAL] Re: [ccp4bb] Lattice-translocation defect (LTD)


EXTERNAL : Real sender is 
owner-ccp...@jiscmail.ac.uk


Hello,

do you have some details about the space group ? Did the integration not miss 
any sports ? I would rather think of an ncs close to crystallographic symmetry, 
or maybe some twinning problem.

I guess these are Pilatus data, can you combine the frames into 1 degree 
oscillations and try Mosflm processing to see how the patterns integrate ?

Greetings

Wim
Le 11/02/2020 à 22:31, Daniele de Sanctis a écrit :
Hi all,

I'm currently dealing with what I think it is a case of LTD (off-origin 
Patterson peak, with vector along w of ~ 7A and electron density map showing a 
"ghost" map shifted by 7 A). I saw there are quite a few cases reported in 
literature  (for example Hare et al, 2006), but what I could not find is how I 
can demodulate the data. Is there any software that can be used for this?

Thank you

Daniele

--
ἀρετή
---
Daniele de Sanctis, PhD

Structural Biology Group
ESRF, Grenoble, France
Tel 33 (0)4 76 88 2869



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

Professeur
Institut de Biologie Structurale (IBS) CIBB
71 avenue des Martyrs / CS 20192
38044 Grenoble Cedex 9, FRANCE
E-mail: wim.burmeis...@ibs.fr
Tel:+33 (0) 457 42 87 41   Fax: +33 (0) 476 20 94 00
website

map

Changement climatique : «Les autres combats n’ont aucun sens si celui-là est 
perdu» (Aurélien Barrau)





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