Re: [ccp4bb] very low R factor for twin refinement

[Robbie Joosten]

Hi Yuri,

The biggest drop I've seen as the result of detwinning is 10% lower R-free. 
Perhaps the detwinning has helped your refinement in othed ways. What R(-free) 
do you get when you take you pdb from the twinned refinement and your input mtz 
file and do 0 cycles of refinememt without detwinning. If you R(-free) is still 
considerably lower than the one from the 'regular' refinememt then you are a 
lucky guy and the twinned refinement just brought you to a lower minimum.

Cheers,
Robbie Joosten

> Date: Thu, 10 Feb 2011 21:36:31 +
> From: ga...@ysbl.york.ac.uk
> Subject: Re: [ccp4bb] very low R factor for twin refinement
> To: CCP4BB@JISCMAIL.AC.UK
> 
> Maximum theoretical drop R/Rfree for perfect twin from 30% is around 25% 
> (i.e. it could go down to 5%). However it could only happen only if twinning 
> is perfect and there is no pseudo rotation parallel to twin operator. 
> Hypothetical case it can happen if you have refined one crystal structure at 
> sufficiently high resolution till (almost convergence) and another crystal is 
> twinned but otherwise perfectly isomorphous to the first crystal and you take 
> coordinates from the first crystal and refine against the second crystal.
> 
> regards
> Garib
> 
> 
> On 10 Feb 2011, at 20:14, Patskovsky Yury wrote:
> 
> > Dear all,
> > 
> > 
> >Twin refinement has yielded  Rwork/Rfree values of about 0.10/0.12 
> > for a nice quality 1.8A dataset (Rmerge 6%, space group I4, twin fractions 
> > 0.6/04) and almost the same R/Rfree (0.095/0.115) for another 1.5A nice 
> > quality data set (Rmerge 6%, space group I4, twin fractions 0.74/0.26). 
> > Refinement of untwinned data resulted in Rfree of ~32% and ~22% 
> > respectively.  REFMAC and PHENIX both have produced the same results and 
> > almost identical R factors, which are suspiciously VERY LOW for this 
> > resolution of data.  Twin refinement in REFMAC has produced exceptional 
> > quality maps even for 1.8A data (they look rather like 1.2A maps)  - I can 
> > not tell the same for PHENIX - maps were looking worse (may be someone has 
> > a better idea why).
> >   Normally twin refinement results in lowering R-factors - say, the 
> > drop in R from 30% (without twin refinement) to 20% (with twin refinement) 
> > would be considered normal, however we can see the drop from 32% to 12%.
> >I wonder if anyone else has experienced similar problems and what 
> > would be the most reasonable explanation for that.
> > 
> > 
> > Thank you
> > 
> > Yury
> > 
> > 
> > 
> > 
> > 
> > Yury Patskovsky, Ph.D.
> > Associate,
> > Dept of Biochemistry
> > Albert Einstein College of Medicine
> > 1300 Morris Park Ave
> > Bronx, NY 10461
> > phone 718-430-2745
> > yu...@medusa.vioc.aecom.yu.edu
  

Re: [ccp4bb] very low R factor for twin refinement

[Robbie Joosten]

Hi Yuri,

The biggest drop I've seen as the result of detwinning is 10% lower R-free. 
Perhaps the detwinning has helped your refinement in othed ways. What R(-free) 
do you get when you take you pdb from the twinned refinement and your input mtz 
file and do 0 cycles of refinememt without detwinning. If you R(-free) is still 
considerably lower than the one from the 'regular' refinememt then you are a 
lucky guy and the twinned refinement just brought you to a lower minimum.

Cheers,
Robbie Joosten

> Date: Thu, 10 Feb 2011 21:36:31 +
> From: ga...@ysbl.york.ac.uk
> Subject: Re: [ccp4bb] very low R factor for twin refinement
> To: CCP4BB@JISCMAIL.AC.UK
> 
> Maximum theoretical drop R/Rfree for perfect twin from 30% is around 25% 
> (i.e. it could go down to 5%). However it could only happen only if twinning 
> is perfect and there is no pseudo rotation parallel to twin operator. 
> Hypothetical case it can happen if you have refined one crystal structure at 
> sufficiently high resolution till (almost convergence) and another crystal is 
> twinned but otherwise perfectly isomorphous to the first crystal and you take 
> coordinates from the first crystal and refine against the second crystal.
> 
> regards
> Garib
> 
> 
> On 10 Feb 2011, at 20:14, Patskovsky Yury wrote:
> 
> > Dear all,
> > 
> > 
> >Twin refinement has yielded  Rwork/Rfree values of about 0.10/0.12 
> > for a nice quality 1.8A dataset (Rmerge 6%, space group I4, twin fractions 
> > 0.6/04) and almost the same R/Rfree (0.095/0.115) for another 1.5A nice 
> > quality data set (Rmerge 6%, space group I4, twin fractions 0.74/0.26). 
> > Refinement of untwinned data resulted in Rfree of ~32% and ~22% 
> > respectively.  REFMAC and PHENIX both have produced the same results and 
> > almost identical R factors, which are suspiciously VERY LOW for this 
> > resolution of data.  Twin refinement in REFMAC has produced exceptional 
> > quality maps even for 1.8A data (they look rather like 1.2A maps)  - I can 
> > not tell the same for PHENIX - maps were looking worse (may be someone has 
> > a better idea why).
> >   Normally twin refinement results in lowering R-factors - say, the 
> > drop in R from 30% (without twin refinement) to 20% (with twin refinement) 
> > would be considered normal, however we can see the drop from 32% to 12%.
> >I wonder if anyone else has experienced similar problems and what 
> > would be the most reasonable explanation for that.
> > 
> > 
> > Thank you
> > 
> > Yury
> > 
> > 
> > 
> > 
> > 
> > Yury Patskovsky, Ph.D.
> > Associate,
> > Dept of Biochemistry
> > Albert Einstein College of Medicine
> > 1300 Morris Park Ave
> > Bronx, NY 10461
> > phone 718-430-2745
> > yu...@medusa.vioc.aecom.yu.edu
  

[ccp4bb] Webinar: Brian Matthews presents "Protein Crystallography: Getting in on the Ground Floor"

[Angela Criswell]

Dear colleagues,

I would like to draw your attention to an upcoming webinar to be presented by 
Brian Matthews, Ph.D. titled "Protein Crystallography: Getting in on the Ground 
Floor". Brian Matthews received both his B.S. (1959) and Ph.D. (1964) from the 
University of Adelaide, completing his studies during a pivotal period in X-ray 
crystallography—at around the same time that the first three-dimensional 
structure of a protein, myoglobin, was determined. "I had the good fortune to 
be a postdoc at the MRC Lab in Cambridge and to participate in the chymotrypsin 
structure determination in David Blow's group." In this webinar, Brian will 
share "a personal recollection of the structural biology at that time."

This webinar is scheduled to occur on Wednesday, February 23rd at 12:00 PM CST 
(10:00 AM PST / 6:00 PM GMT). You can find more information, including a 
registration link at:  http://www.rigaku.com/protein/webinars.html

Best regards,
Angela

-- 
Angela R. Criswell, Ph. D.
Rigaku Americas Corporation
9009 New Trails Drive
The Woodlands, TX 77381 USA
Ph: +1 281 362 2300  ext. 216
Fax: +1 281 364 3628
Email: angela.crisw...@rigaku.com
URL: http://www.rigaku.com

Re: [ccp4bb] very low R factor for twin refinement

[Garib N Murshudov]

Dear Yury

I looked at this data and such drop of R/Rfree when you switch from non-twin to 
twin in the presence of twinning is expected.

When you said that electron density is too good I was afraid that there is 
strong bias. But at least for the case from the pdb (2gl5) you can see a lot of 
features that are not in the model (double conformations, waters, end even some 
unexplained ligands etc).  In general, maps, especially unexplained parts are 
best indicators of bias towards the errors in the model.



Regards
Garib


On 10 Feb 2011, at 22:06, Patskovsky Yury wrote:

> Thanks, Garib
> 
> It might be the case.
> As a matter of fact, you are welcome to look at the original data here
> 
> http://www.rcsb.org/pdb/explore/explore.do?structureId=2GL5
> 
> The data turned out to be twinned ( I also have other datasets - all with 
> certain degree of twinning) and now I am trying to re-refine and then 
> re-submit the more correct structure to the PDB.
> 
> 
> 
> Yury
> 
> On Thu, 10 Feb 2011, Garib N Murshudov wrote:
> 
>> Maximum theoretical drop R/Rfree for perfect twin from 30% is around 25% 
>> (i.e. it could go down to 5%). However it could only happen only if twinning 
>> is perfect and there is no pseudo rotation parallel to twin operator.
>> Hypothetical case it can happen if you have refined one crystal structure at 
>> sufficiently high resolution till (almost convergence) and another crystal 
>> is twinned but otherwise perfectly isomorphous to the first crystal and you 
>> take coordinates from the first crystal and refine against the second 
>> crystal.
>> 
>> regards
>> Garib
>> 
>> 
>> On 10 Feb 2011, at 20:14, Patskovsky Yury wrote:
>> 
>>> Dear all,
>>> 
>>> 
>>>   Twin refinement has yielded  Rwork/Rfree values of about 0.10/0.12 
>>> for a nice quality 1.8A dataset (Rmerge 6%, space group I4, twin fractions 
>>> 0.6/04) and almost the same R/Rfree (0.095/0.115) for another 1.5A nice 
>>> quality data set (Rmerge 6%, space group I4, twin fractions 0.74/0.26). 
>>> Refinement of untwinned data resulted in Rfree of ~32% and ~22% 
>>> respectively.  REFMAC and PHENIX both have produced the same results and 
>>> almost identical R factors, which are suspiciously VERY LOW for this 
>>> resolution of data.  Twin refinement in REFMAC has produced exceptional 
>>> quality maps even for 1.8A data (they look rather like 1.2A maps)  - I can 
>>> not tell the same for PHENIX - maps were looking worse (may be someone has 
>>> a better idea why).
>>>  Normally twin refinement results in lowering R-factors - say, the drop 
>>> in R from 30% (without twin refinement) to 20% (with twin refinement) would 
>>> be considered normal, however we can see the drop from 32% to 12%.
>>>   I wonder if anyone else has experienced similar problems and what 
>>> would be the most reasonable explanation for that.
>>> 
>>> 
>>> Thank you
>>> 
>>> Yury
>>> 
>>> 
>>> 
>>> 
>>> 
>>> Yury Patskovsky, Ph.D.
>>> Associate,
>>> Dept of Biochemistry
>>> Albert Einstein College of Medicine
>>> 1300 Morris Park Ave
>>> Bronx, NY 10461
>>> phone 718-430-2745
>>> yu...@medusa.vioc.aecom.yu.edu
>> 
>> 

Re: [ccp4bb] very low R factor for twin refinement

[Pavel Afonine]

Hi Yury,

- make sure you correctly generated free-R flags, so there is no cross-talk
between test/work sets. Illustration: see page number 123 here:
http://cci.lbl.gov/~afonine/for_ccp4/PavelAfonine_PHENIX.pdf

- compare regular vs missing-fobs-filled maps. See pages188-190 here:
http://cci.lbl.gov/~afonine/for_ccp4/PavelAfonine_PHENIX.pdf
and previous discussion on phenixbb about this:
http://www.phenix-online.org/pipermail/phenixbb/2009-August/002315.html

Pavel.


   Twin refinement has yielded  Rwork/Rfree values of about 0.10/0.12
> for a nice quality 1.8A dataset (Rmerge 6%, space group I4, twin fractions
> 0.6/04) and almost the same R/Rfree (0.095/0.115) for another 1.5A nice
> quality data set (Rmerge 6%, space group I4, twin fractions 0.74/0.26).
> Refinement of untwinned data resulted in Rfree of ~32% and ~22%
> respectively.  REFMAC and PHENIX both have produced the same results and
> almost identical R factors, which are suspiciously VERY LOW for this
> resolution of data.  Twin refinement in REFMAC has produced exceptional
> quality maps even for 1.8A data (they look rather like 1.2A maps)  - I can
> not tell the same for PHENIX - maps were looking worse (may be someone has a
> better idea why).
>   Normally twin refinement results in lowering R-factors - say, the
> drop in R from 30% (without twin refinement) to 20% (with twin refinement)
> would be considered normal, however we can see the drop from 32% to 12%.
>I wonder if anyone else has experienced similar problems and what
> would be the most reasonable explanation for that.
>
>

Re: [ccp4bb] very low R factor for twin refinement

[Garib N Murshudov]

Just a warning: When doing twin refinement (and in general) you should always 
use original data for refinement. After refinement data may be different (in 
case of twin de-twinned).

I will have a look as soon as I have some time.

regards
Garib


On 10 Feb 2011, at 22:06, Patskovsky Yury wrote:

> Thanks, Garib
> 
> It might be the case.
> As a matter of fact, you are welcome to look at the original data here
> 
> http://www.rcsb.org/pdb/explore/explore.do?structureId=2GL5
> 
> The data turned out to be twinned ( I also have other datasets - all with 
> certain degree of twinning) and now I am trying to re-refine and then 
> re-submit the more correct structure to the PDB.
> 
> 
> 
> Yury
> 
> On Thu, 10 Feb 2011, Garib N Murshudov wrote:
> 
>> Maximum theoretical drop R/Rfree for perfect twin from 30% is around 25% 
>> (i.e. it could go down to 5%). However it could only happen only if twinning 
>> is perfect and there is no pseudo rotation parallel to twin operator.
>> Hypothetical case it can happen if you have refined one crystal structure at 
>> sufficiently high resolution till (almost convergence) and another crystal 
>> is twinned but otherwise perfectly isomorphous to the first crystal and you 
>> take coordinates from the first crystal and refine against the second 
>> crystal.
>> 
>> regards
>> Garib
>> 
>> 
>> On 10 Feb 2011, at 20:14, Patskovsky Yury wrote:
>> 
>>> Dear all,
>>> 
>>> 
>>>   Twin refinement has yielded  Rwork/Rfree values of about 0.10/0.12 
>>> for a nice quality 1.8A dataset (Rmerge 6%, space group I4, twin fractions 
>>> 0.6/04) and almost the same R/Rfree (0.095/0.115) for another 1.5A nice 
>>> quality data set (Rmerge 6%, space group I4, twin fractions 0.74/0.26). 
>>> Refinement of untwinned data resulted in Rfree of ~32% and ~22% 
>>> respectively.  REFMAC and PHENIX both have produced the same results and 
>>> almost identical R factors, which are suspiciously VERY LOW for this 
>>> resolution of data.  Twin refinement in REFMAC has produced exceptional 
>>> quality maps even for 1.8A data (they look rather like 1.2A maps)  - I can 
>>> not tell the same for PHENIX - maps were looking worse (may be someone has 
>>> a better idea why).
>>>  Normally twin refinement results in lowering R-factors - say, the drop 
>>> in R from 30% (without twin refinement) to 20% (with twin refinement) would 
>>> be considered normal, however we can see the drop from 32% to 12%.
>>>   I wonder if anyone else has experienced similar problems and what 
>>> would be the most reasonable explanation for that.
>>> 
>>> 
>>> Thank you
>>> 
>>> Yury
>>> 
>>> 
>>> 
>>> 
>>> 
>>> Yury Patskovsky, Ph.D.
>>> Associate,
>>> Dept of Biochemistry
>>> Albert Einstein College of Medicine
>>> 1300 Morris Park Ave
>>> Bronx, NY 10461
>>> phone 718-430-2745
>>> yu...@medusa.vioc.aecom.yu.edu
>> 
>> 

Re: [ccp4bb] very low R factor for twin refinement

[Patskovsky Yury]

Thanks, Garib

It might be the case.
As a matter of fact, you are welcome to look at the original data here

http://www.rcsb.org/pdb/explore/explore.do?structureId=2GL5

The data turned out to be twinned ( I also have other datasets - all 
with certain degree of twinning) and now I am trying to re-refine and 
then re-submit the more correct structure to the PDB.




Yury

On Thu, 10 Feb 2011, Garib N Murshudov wrote:


Maximum theoretical drop R/Rfree for perfect twin from 30% is around 25% (i.e. 
it could go down to 5%). However it could only happen only if twinning is 
perfect and there is no pseudo rotation parallel to twin operator.
Hypothetical case it can happen if you have refined one crystal structure at 
sufficiently high resolution till (almost convergence) and another crystal is 
twinned but otherwise perfectly isomorphous to the first crystal and you take 
coordinates from the first crystal and refine against the second crystal.

regards
Garib


On 10 Feb 2011, at 20:14, Patskovsky Yury wrote:


Dear all,


   Twin refinement has yielded  Rwork/Rfree values of about 0.10/0.12 for a 
nice quality 1.8A dataset (Rmerge 6%, space group I4, twin fractions 0.6/04) 
and almost the same R/Rfree (0.095/0.115) for another 1.5A nice quality data 
set (Rmerge 6%, space group I4, twin fractions 0.74/0.26). Refinement of 
untwinned data resulted in Rfree of ~32% and ~22% respectively.  REFMAC and 
PHENIX both have produced the same results and almost identical R factors, 
which are suspiciously VERY LOW for this resolution of data.  Twin refinement 
in REFMAC has produced exceptional quality maps even for 1.8A data (they look 
rather like 1.2A maps)  - I can not tell the same for PHENIX - maps were 
looking worse (may be someone has a better idea why).
  Normally twin refinement results in lowering R-factors - say, the drop in 
R from 30% (without twin refinement) to 20% (with twin refinement) would be 
considered normal, however we can see the drop from 32% to 12%.
   I wonder if anyone else has experienced similar problems and what would 
be the most reasonable explanation for that.


Thank you

Yury





Yury Patskovsky, Ph.D.
Associate,
Dept of Biochemistry
Albert Einstein College of Medicine
1300 Morris Park Ave
Bronx, NY 10461
phone 718-430-2745
yu...@medusa.vioc.aecom.yu.edu

Re: [ccp4bb] very low R factor for twin refinement

[Tommi Kajander]

you didnt by any chance mix up Rfree and the rest of the data flags at  
any point...??? (that wouldnt probably explain the low Rfree value  
though...)
unless you then mix them up again mid -way through... check your  
flags? (1 or 0)


tommi

On Feb 10, 2011, at 11:36 PM, Garib N Murshudov wrote:

Maximum theoretical drop R/Rfree for perfect twin from 30% is around  
25% (i.e. it could go down to 5%). However it could only happen only  
if twinning is perfect and there is no pseudo rotation parallel to  
twin operator.
Hypothetical case it can happen if you have refined one crystal  
structure at sufficiently high resolution till (almost convergence)  
and another crystal is twinned but otherwise perfectly isomorphous  
to the first crystal and you take coordinates from the first crystal  
and refine against the second crystal.


regards
Garib


On 10 Feb 2011, at 20:14, Patskovsky Yury wrote:


Dear all,


  Twin refinement has yielded  Rwork/Rfree values of about  
0.10/0.12 for a nice quality 1.8A dataset (Rmerge 6%, space group  
I4, twin fractions 0.6/04) and almost the same R/Rfree  
(0.095/0.115) for another 1.5A nice quality data set (Rmerge 6%,  
space group I4, twin fractions 0.74/0.26). Refinement of untwinned  
data resulted in Rfree of ~32% and ~22% respectively.  REFMAC and  
PHENIX both have produced the same results and almost identical R  
factors, which are suspiciously VERY LOW for this resolution of  
data.  Twin refinement in REFMAC has produced exceptional quality  
maps even for 1.8A data (they look rather like 1.2A maps)  - I can  
not tell the same for PHENIX - maps were looking worse (may be  
someone has a better idea why).
 Normally twin refinement results in lowering R-factors - say,  
the drop in R from 30% (without twin refinement) to 20% (with twin  
refinement) would be considered normal, however we can see the drop  
from 32% to 12%.
  I wonder if anyone else has experienced similar problems and  
what would be the most reasonable explanation for that.



Thank you

Yury





Yury Patskovsky, Ph.D.
Associate,
Dept of Biochemistry
Albert Einstein College of Medicine
1300 Morris Park Ave
Bronx, NY 10461
phone 718-430-2745
yu...@medusa.vioc.aecom.yu.edu




Tommi Kajander, Ph.D.
Structural Biology and Biophysics
Institute of Biotechnology
University of Helsinki
Viikinkaari 1
(P.O. Box 65)
00014 Helsinki
Finland
p. +358-9-19158903
tommi.kajan...@helsinki.fi

Re: [ccp4bb] very low R factor for twin refinement

[Garib N Murshudov]

Maximum theoretical drop R/Rfree for perfect twin from 30% is around 25% (i.e. 
it could go down to 5%). However it could only happen only if twinning is 
perfect and there is no pseudo rotation parallel to twin operator. 
Hypothetical case it can happen if you have refined one crystal structure at 
sufficiently high resolution till (almost convergence) and another crystal is 
twinned but otherwise perfectly isomorphous to the first crystal and you take 
coordinates from the first crystal and refine against the second crystal.

regards
Garib


On 10 Feb 2011, at 20:14, Patskovsky Yury wrote:

> Dear all,
> 
> 
>Twin refinement has yielded  Rwork/Rfree values of about 0.10/0.12 for 
> a nice quality 1.8A dataset (Rmerge 6%, space group I4, twin fractions 
> 0.6/04) and almost the same R/Rfree (0.095/0.115) for another 1.5A nice 
> quality data set (Rmerge 6%, space group I4, twin fractions 0.74/0.26). 
> Refinement of untwinned data resulted in Rfree of ~32% and ~22% respectively. 
>  REFMAC and PHENIX both have produced the same results and almost identical R 
> factors, which are suspiciously VERY LOW for this resolution of data.  Twin 
> refinement in REFMAC has produced exceptional quality maps even for 1.8A data 
> (they look rather like 1.2A maps)  - I can not tell the same for PHENIX - 
> maps were looking worse (may be someone has a better idea why).
>   Normally twin refinement results in lowering R-factors - say, the drop 
> in R from 30% (without twin refinement) to 20% (with twin refinement) would 
> be considered normal, however we can see the drop from 32% to 12%.
>I wonder if anyone else has experienced similar problems and what 
> would be the most reasonable explanation for that.
> 
> 
> Thank you
> 
> Yury
> 
> 
> 
> 
> 
> Yury Patskovsky, Ph.D.
> Associate,
> Dept of Biochemistry
> Albert Einstein College of Medicine
> 1300 Morris Park Ave
> Bronx, NY 10461
> phone 718-430-2745
> yu...@medusa.vioc.aecom.yu.edu

[ccp4bb] FSEARCH question

[Peter Grey]

Dear CCP4  users,

I am trying to use FSEARCH to position a SAXS envelop in an attempt to solve
a molecular replacement case.
Could you explain what the x,y,z values in the output stand for. Do they
represent the position of the center of mass of the molecule in the unit
cell ?

I would be grateful if you could shed some light on the matter,


Peter

[ccp4bb] Postdoctoral position at Auckland

[Alok Mitra]

Dear all:
A postdoctoral position is available in the Mitra Lab at the School of 
Biological Sciences at University of Auckland to investigate 3-dimensional 
structure/function of two membrane proteins involved in cation and solute 
transport using primarily X-ray crystallography, but also electron 
crystallography when appropriate. Purified protein, either produced by 
recombinant expression or from natural sources are now available in amounts 
sufficient for extensive crystal trials, which are ongoing.
The ideal candidate should have a PhD in biochemistry, biophysics or structural 
biology preferably with an experience in membrane proteins. Previous experience 
in X-ray crystallography and the use of biophysical techniques for membrane 
proteins would be an advantage, so also knowledge of basic molecular biology 
techniques and protein expression both from bacteria and yeast. The position is 
initially for 2 years and 4 months with a good possibility for extension. The 
salary will be commensurate with experience.
Our group is part of the Structural Biology Section in the School comprising of 
5 other groups. Modern in-house facilities for carrying out all aspects of 
protein crystallography are available (directed by Prof. Ted baker) as part of 
the core facility.  The X-ray suite comprises of a Rigaku MicroMax(TM) - 007HR 
rotation anode equipped with 2x Mar345dtb detectors & 2x Oxford Cryosystems 
Cobra units. The 3D crystallization suite comprises of a Cartesian HONEYBEE 
Benchtop Crystallisation System with Perkin Elmer MultiPROBE II HT.  We also 
have direct access to the Australian Synchotron facility. We have a modern 
cryo-EM unit comprising of a CM12, a Tecnai12 equipped with a 2K2K GATAN CCD 
camera and a TecnaiF20 with energy filter to arrive mid 2011 and all ancillary 
equipments for cryo work. A 600MHZ NMR machine is also available, which is 
housed in the Chemistry Dept. Details of other facilities available within the 
school may be found from http://www.sbs.auckland.ac.nz/uoa/.
Auckland has been voted the 5th most liveable city in the world. The city is 
dotted with numerous parks and beaches and has a temperate climate. New Zealand 
is a haven for nature lovers.
Informal enquires can be made to Alok K Mitra (a.mi...@auckland.ac.nz).  The 
position will be open until filled.

[ccp4bb] very low R factor for twin refinement

[Patskovsky Yury]

Dear all,


Twin refinement has yielded  Rwork/Rfree values of about 0.10/0.12 
for a nice quality 1.8A dataset (Rmerge 6%, space group I4, twin fractions 
0.6/04) and almost the same R/Rfree (0.095/0.115) for another 1.5A nice 
quality data set (Rmerge 6%, space group I4, twin fractions 0.74/0.26). 
Refinement of untwinned data resulted in Rfree of ~32% and ~22% 
respectively.  REFMAC and PHENIX both have produced the same results and 
almost identical R factors, which are suspiciously VERY LOW for this 
resolution of data.  Twin refinement in REFMAC has produced exceptional 
quality maps even for 1.8A data (they look rather like 1.2A maps)  - I can 
not tell the same for PHENIX - maps were looking worse (may be someone has 
a better idea why).
   Normally twin refinement results in lowering R-factors - say, the 
drop in R from 30% (without twin refinement) to 20% (with twin refinement) 
would be considered normal, however we can see the drop from 32% to 12%.
I wonder if anyone else has experienced similar problems and what 
would be the most reasonable explanation for that.



Thank you

Yury





Yury Patskovsky, Ph.D.
Associate,
Dept of Biochemistry
Albert Einstein College of Medicine
1300 Morris Park Ave
Bronx, NY 10461
phone 718-430-2745
yu...@medusa.vioc.aecom.yu.edu

Re: [ccp4bb] First images of proteins and viruses caught with an X-ray laser

[James Holton]

Our main reason for thinking that the nanocrystals were not twinned is 
because the "regular" crystals of the same form, grown under very 
similar conditions were not twinned (1jb0).  The space group, unit cell, 
and indeed the structure in the ASU were all the same.  In fact, I don't 
think anyone has grown twinned crystals of this protein?


But, to address Colin's question:

In Figure 2 of the Nature paper we showed some "Shrinkwrap" 
reconstructions of the crystal shape from the spot shape.  For those who 
don't know Shrinnkwrap, it works a lot like "dm" or "solomon", except 
you have 99.9% solvent content (the area outside the crystal), so it is 
fairly robust.  In all the cases I have seen, there is only one blob in 
the shape reconstruction, indicating a single (non-twinned) crystal.  In 
fact, many of them look like a hexagonal prismatic solid, which is what 
samples of these crystals look like under EM.  This was done for several 
images, but certainly not all of them.  I don't think any single human 
being has looked at all of the images.


However, maybe Colin would be interested in the "double hits"?  The 
nanocrystals were diluted to avoid them, but every so often two crystals 
are caught in a single shot, and you see two lattices on the detector.  
Autoindexing generally fails in these cases, but the distance between 
the two crystallites should be apparent as a modulation of the fringes.


Colin, maybe you could ask Henry Chapman or John Spence about that?

-James Holton
MAD Scientist

On 2/10/2011 3:14 AM, Colin Nave wrote:


James

All great stuff.

ØThe individual crystals were not twinned (or at least I would be VERY 
surprised if they were)


Can this be tested? The spots should show fringes around them 
corresponding to the shape transform of the nano-crystal. If twinned, 
the intensities of the fringes should be modified due to the contrast 
between the two types of domain for the particular reflection. If it 
worked, one would be equivalent to grinding up the crystal using the 
coherent beam.


Probably difficult as would have to be done independently for each 
crystal.


Colin

*From:*CCP4 bulletin board [mailto:CCP4BB@JISCMAIL.AC.UK] *On Behalf 
Of *James Holton

*Sent:* 10 February 2011 01:20
*To:* CCP4BB@JISCMAIL.AC.UK
*Subject:* Re: [ccp4bb] First images of proteins and viruses caught 
with an X-ray laser


The twin fraction for REFMAC was exactly 0.5.  The individual crystals 
were not twinned (or at least I would be VERY surprised if they were), 
but they do belong to the space group P63, and autoindexing will give 
you one of the two possible axis conventions at random.  So, of the 
15445 crystals that were merged, there were about ~7700 indexed one 
way and 7700 indexed the other way.  It is hard to tell exactly.  
Mergeing the data "blindly" (which we did) will result in a "twinned 
data set" where each merged h,k,l has an equal contribution from 
k,h,-l.  This was all discussed in the data processing paper PMID: 
20389587.


Obviously, there are a number of things you can think of for resolving 
this indexing ambiguity and removing the "twinning effect", and in 
fact, this is what I thought was really cool about this particular 
kind of data collection and why I encouraged the people doing all the 
work (the other 87 authors) to go ahead with a "twinnable" space 
group.  On the first pass, you can just take advantage of TWIN 
refinement in REFMAC, but in the future, when the indexing improves 
(probably using post-refinement) it may be possible to de-twin a 
crystal system that actually suffers from "real" twinning.  That is, a 
"twin domain" cannot be smaller than a mosaic block (otherwise the 
h,k,l and -k,h,-l structure factors would add as phased Fs, not 
|F|^2).  So, since nanocrystals are essentially single mosaic blocks 
(smaller than the coherence length of the beam), you could take your 
twinned crystals and either grind them up or re-optimize for smaller 
crystals (counterintuitive!), and then resolve the twin domains 
"manually".  Sort of like Louis Pasteur and his tartaric acid crystals.


Anyway, I thought that was a cool idea, but like so many other cool 
things, it had to be cut from the Nature paper.  Admittedly, the 
problem has not actually been solved yet.  This is why we used REFMAC 
in TWIN mode.


-James Holton
MAD Scientist

On Wed, Feb 9, 2011 at 3:29 PM, Jon Schuermann > wrote:


According to the paper, the data was refined in REFMAC in 'twin mode' 
which, I believe, calculates the R-factor using a  non-conventional 
R-factor equation which usually lower than the conventional R-factor. 
I believe this is dependent on the twin fraction which wasn't 
mentioned in the paper (or supplementary info) unless I missed it.


Jon


--
Jonathan P. Schuermann, Ph. D.
Beamline Scientist
NE-CAT, Building 436E
Advanced Photon Source (APS)
Argonne National Laboratory
9700 South Cass Avenue
Argonne, IL 60439
  
email:schue...@anl.gov  

Re: [ccp4bb] First images of proteins and viruses caught with an X-ray laser

[James Holton]

I hope everyone understands that I am not the corresponding author on 
this paper!  That is Henry Chapman.  He, and most of the other authors 
have been slaving away on this problem for most of their professional 
careers.  To be honest, I never did think it was going to work, and I 
was absolutely "gobsmacked" as they say when I heard they got spots 
going all the way out to the edge of the detector!


However, since my involvement was mostly about the data processing and 
refinement, I think Henry won't mind if I respond to a few of the 
questions posted on the BB about it.


Hudel:

Hmm.  Looks like our paper didn't report the R/Rfree for the "control" 
data set.  Crap.  Sorry about that.  For the "record", it was 
0.27/0.30.  These "control" data were collected from a large and 
cryo-cooled photosystem I crystal at ALS 8.2.2, truncated to 8.5 A and 
artificially "twinned" by averaging F(h,k,l)^2 and F(k,h,-l)^2, then 
subjected to the same rigid-body REFMAC refinement as the LCLS data.  
Exactly the same Rfree flags were used for both cases, and these were 
chosen in the point group 622 so that no "twin mates" were in both the 
working and test sets.


And before you ask, no, I have no idea why the "control" data were 
"worse".  However, I can say that they were not all that different from 
the LCLS data.  The maps (which were shown) don't really look all that 
different (IMHO), and the R-iso between the ALS and LCLS datasets was 
22% (also shown).  I thought the latter was amazing agreement 
considering that the ALS crystal was cryo-cooled, the LCLS crystals were 
at room temperature and the lowest R-iso between lysozyme datasets in 
the PDB collected at room temp vs cryo is 20% (2hu1 vs 2epe).  Yes, I 
checked all 10,000 lysozyme-lysozyme pairings.


I'm afraid I can't directly address Hudel's Rtwin = 0.7*Rnormal rule 
because de-twinned LCLS data are not yet available and I don't have the 
untwinned ALS data on my computers.  The reason I was "impressed" was 
because data in this resolution range generally don't fit very well to 
molecular models (have a look at R/Rfree in the infinity-9A bin of your 
last structure).  So, I was surprised to see that R/Rfree was "okay".  I 
guess I am just a pessimist.  All I can really say is that we used 
Refmac_5.6.0076, and the data are deposited under 3pcq if anyone wants 
to play with it.


-James Holton
MAD Scientist

On 2/9/2011 10:28 PM, Hudel Luecke wrote:

James,

I must admit not (yet) having read the "data processing paper PMID: 20389587" nor the 88 
author paper.  Nevertheless, I would like to comment on your remark "Personally, I was quite 
impressed by how good the R factors were, all things considered."

If I am not mistaken, perfectly twinned intensity data, such as you seem to 
have been dealing with, will generate data sets with compressed dynamic ranges, 
which in turn means that crystallographic R factors computed based on such data 
sets will be lower by roughly sqrt(2) or 1.4 compared to a non-twinned data 
set.  So a somewhat iffy R(cryst) of 30% would look quite nice (21.4%) when 
computed on a hemihedrally twinned data set.  But maybe this is all discussed 
in your paper...


Cheers, Hudel

[ccp4bb] SLS MX Beamline Proposals: Deadline is approaching !

[Vincent Olieric]

===
SYNCHROTRON BEAM TIME FOR MACROMOLECULAR CRYSTALLOGRAPHY AT SLS
===

Proposal application deadline: Tuesday, February 15, 2011

Periods:
May 1, 2011 - August 31, 2011 (Normal proposals)
May 1, 2011 - April 30, 2013 (Long-term proposals)

Proposal submission:
http://www.psi.ch/sls/px-beamlines-call-for-proposals

Travel support:
http://www.psi.ch/useroffice/sls-elisa

News:
- In-situ X-ray diffraction screening available at X06DA (with any
SBS-format plates and at any time during a shift)
- Possibility to combine beamtime at X06SA and X06DA, please indicate
your shift requests in the proposal (only one proposal is needed on
either X06SA or X06DA)
- X10SA accepts proposals for combined single crystal diffraction and
UV-Vis, fluorescence and Raman spectroscopy studies (see R.Owen, et
al., JSR, 2009, 16, 173)
- Automatic sample changer (IRELEC CATS) with SPINE pucks, vials, and
caps (MOLECULAR DIMENSIONS) available at all beamline (except on the
MD2). Only day time (9:00~23:00) operation is supported.
- PSI DUO application for iphone:
http://itunes.apple.com/ch/app/psi-duo/id375328818?mt=8


X06SA Beamline features (http://www.psi.ch/sls/pxi):
- Undulator beamline with flux of 2x10E12 photons/sec at 12.4 keV (1Å)
and fully tunable from 6.0 to 17.5 keV (2.07 - 0.71 Å)
- Focused beam size and divergency: HRD - 85x10 microns and 0.35x0.06
mrad; MD2 - 25x5 microns and 0.5x0.4 mrad
- PILATUS 6M pixel detector at the High Resolution Diffractometer,
allowing continuous, fine phi-sliced data acquisition (12.5 frames per
second) with 20 bit dynamic range (see http://pilatus.web.psi.ch/ or
www.dectris.com for further informations)
- MAR225 CCD at Micro-Diffractometer MD2, allowing data collection
with a focussed beam size of 25x5 micrometers, and down to 10x5
micrometer using an aperture.

X06DA Beamline features (http://www.psi.ch/sls/pxiii):
- In-situ X-ray diffraction screening capabilities (available at any
time during a shift, all SBS-format plates supported)
- Super-bending magnet beamline with flux of 5x10E11 photons/sec at
12.4 keV (1Å) and fully tunable from 6.0 to 17.5 keV (2.07 - 0.71 Å)
- Focused beam size and divergency: 80x45 microns and 2x0.5 mrad (with
possibility to reduce horizontal divergency to 0.4 mrad)
- Mini-hutch design for fast manual mounting
- MAR225 CCD detector

X10SA Beamline features (http://www.psi.ch/sls/pxii):
- In-situ on-axis single crystal UV-Vis, fluorescence, Raman and
Resonance Raman microspectrophotometer
- Undulator beamline with flux of 2x10E12 photons/sec at 12.4 keV (1Å)
and fully tunable from 6.0 to 20 keV (2.07 - 0.62 Å)
- Focused beam size and divergency: 50x10 microns and 0.6x0.1 mrad
- PILATUS 6M pixel detector

Best regards,

The MX group at SLS

--
Dr Vincent Olieric
Beamline Scientist
Swiss Light Source at Paul Scherrer Institut
WSLA/218
5232 Villigen PSI - Switzerland
+41 (0)56 310 5233

Re: [ccp4bb] exploding hydrogens during real space refinement in coot

[Christian Roth]

That happens also for me with all hydrogens, regardless if they belongs to a 
ligand or protein.
Best solution for me is stripping all hydrogens out of the pdb. Manipulating 
the file in coot and  readding them and start of the renfinement.

Christian

Am Donnerstag 10 Februar 2011 17:04:18 schrieben Sie:
> When I real space refine my ligand the hydrogens fly away. The ligand pdb
>  file and cif file were generated with progdr. I'm using Coot 0.6.2-pre-1.
> 
> Kendall
> 

Re: [ccp4bb] exploding hydrogens during real space refinement in coot

[Folmer Fredslund]

Dear Kendall

2011/2/10 Kendall Nettles :
> When I real space refine my ligand the hydrogens fly away. The ligand pdb 
> file and cif file were generated with progdr. I'm using Coot 0.6.2-pre-1.
>
> Kendall
>

This could be a problem with the naming conventions of pdb versions. I
don't remember which version Coot uses, but here is an old thread on
the subject http://www.mail-archive.com/coot@jiscmail.ac.uk/msg00405.html

Best regards,
-- 
Folmer Fredslund

[ccp4bb] exploding hydrogens during real space refinement in coot

[Kendall Nettles]

When I real space refine my ligand the hydrogens fly away. The ligand pdb file 
and cif file were generated with progdr. I'm using Coot 0.6.2-pre-1. 

Kendall 

Re: [ccp4bb] First images of proteins and viruses caught with an X-ray laser

[Jacob Keller]

Would it be true that the anomalous differences could not be measured
in these types of datasets, because one would not know which
Friedel/Bivoet reflection one is measuring in a given frame? Perhaps,
given anomalous signal, there would be a way to tease out which
orientation one was looking at from the correlations of the
signs/magnitudes of anomalous-scattering-induced deviations from the
mean intensities (derived from the whole dataset) for all of the
relections observed in each frame? I guess this might also detwin the
data?

JPK

On Thu, Feb 10, 2011 at 7:17 AM, Anastassis Perrakis  wrote:
>>
>> Anyway, I thought that was a cool idea, but like so many other cool
>> things, it had to be cut from the Nature paper.  Admittedly, the problem has
>> not actually been solved yet.  This is why we used REFMAC in TWIN mode.
>
> Is that a hint on the:
>
> a. wisdom of the editor
> b. wisdom of 'the third referee'
> c. wisdom of the dogma 'five years of eight eight lifes in 2000 words'
> d. All of the above
>
> ;-)
>
> A.
>



-- 
***
Jacob Pearson Keller
Northwestern University
Medical Scientist Training Program
cel: 773.608.9185
email: j-kell...@northwestern.edu
***

Re: [ccp4bb] ctruncate - FP=0?

[Phil Evans]

This does sound like a bug. If one of h,k,l are zero then the reflection is 
centric in P222 so the truncation will be different

I just ran a test on an orthorhombic dataset (P212121) of mine and I do indeed 
see some strange F=0, sigF>0  reflections, but  in Charles Ballard's 
development version which I've been testing they are OK.

I think you'll have to send your data to Charles Ballard (c...@ccp4.ac.uk) who 
is working on ctruncate (or Charles I can send you mine if you like)

Phil

On 10 Feb 2011, at 13:50, Ed Pozharski wrote:

> Excellent point and no, these are not missing reflections.  SigF is not
> zero.  Also, if I am not mistaken, missing reflections in the MTZ format
> are recorded as NaN.
> 
> Ed.
> 
> On Thu, 2011-02-10 at 12:17 +, Eleanor Dodson wrote:
>> Are you sure these are real FP=0 or reflections which werent measured 
>> but have been added for completeness of the h k l list.
>> The check is whether the SigF is also 0.00 - in that case they are 
>> genuinely missing..
>> 
>> Eleanor
>> 
>> On 02/09/2011 11:34 PM, Ed Pozharski wrote:
>>> I observe under some conditions that ctruncate sets some reflections
>>> amplitudes to zero.  AFAIU, this should not be happening as even
>>> negative intensities (there are none in this particular dataset) should
>>> produce FP>0 upon truncation.
>>> 
>>> 66 out of ~23000 reflections are zeros after ctruncate is applied.
>>> Nothing obvious comes up upon inspecting the corresponding hkl's, except
>>> that one and only one is always zero (sg is P21212, so these are not
>>> systematic absences).  One curious thing is that the I/sigma for these
>>> reflections is close to the average I/sigma in the highest resolution
>>> range (but it varies and these reflections are in all resolution
>>> ranges).
>>> 
>>> A bug?
>>> 
>> 
> 
> -- 
> Edwin Pozharski, PhD, Assistant Professor
> University of Maryland, Baltimore
> --
> When the Way is forgotten duty and justice appear;
> Then knowledge and wisdom are born along with hypocrisy.
> When harmonious relationships dissolve then respect and devotion arise;
> When a nation falls to chaos then loyalty and patriotism are born.
> --   / Lao Tse /

Re: [ccp4bb] ctruncate - FP=0?

[Ed Pozharski]

Excellent point and no, these are not missing reflections.  SigF is not
zero.  Also, if I am not mistaken, missing reflections in the MTZ format
are recorded as NaN.

Ed.

On Thu, 2011-02-10 at 12:17 +, Eleanor Dodson wrote:
> Are you sure these are real FP=0 or reflections which werent measured 
> but have been added for completeness of the h k l list.
> The check is whether the SigF is also 0.00 - in that case they are 
> genuinely missing..
> 
> Eleanor
> 
> On 02/09/2011 11:34 PM, Ed Pozharski wrote:
> > I observe under some conditions that ctruncate sets some reflections
> > amplitudes to zero.  AFAIU, this should not be happening as even
> > negative intensities (there are none in this particular dataset) should
> > produce FP>0 upon truncation.
> >
> > 66 out of ~23000 reflections are zeros after ctruncate is applied.
> > Nothing obvious comes up upon inspecting the corresponding hkl's, except
> > that one and only one is always zero (sg is P21212, so these are not
> > systematic absences).  One curious thing is that the I/sigma for these
> > reflections is close to the average I/sigma in the highest resolution
> > range (but it varies and these reflections are in all resolution
> > ranges).
> >
> > A bug?
> >
> 

-- 
Edwin Pozharski, PhD, Assistant Professor
University of Maryland, Baltimore
--
When the Way is forgotten duty and justice appear;
Then knowledge and wisdom are born along with hypocrisy.
When harmonious relationships dissolve then respect and devotion arise;
When a nation falls to chaos then loyalty and patriotism are born.
--   / Lao Tse /

Re: [ccp4bb] First images of proteins and viruses caught with an X-ray laser

[Anastassis Perrakis]

Anyway, I thought that was a cool idea, but like so many other cool  
things, it had to be cut from the Nature paper.  Admittedly, the  
problem has not actually been solved yet.  This is why we used  
REFMAC in TWIN mode.


Is that a hint on the:

a. wisdom of the editor
b. wisdom of 'the third referee'
c. wisdom of the dogma 'five years of eight eight lifes in 2000 words'
d. All of the above

;-)

A.

[ccp4bb] SUMMARY [ccp4bb] group defs for refmac rigid body

[Tim Gruene]

Hello,

I received two private emails answering my question. Both pointed out that the
residues are allowed to go beyond the actual residue numbers in the PDB file, so
that specifying e.g. residues 1 through  works just like a wildcard.

Cheers, Tim

On Wed, Feb 09, 2011 at 12:26:34PM +0100, Tim Gruene wrote:
> Dear all,
> 
> is there a short cut to define an entire chain as one group in the rigid body
> refinement of refmac, or do I have to explicitly name the starting and ending
> residues - that's a little tedious with several domains, and using a wild card
> would come very handy.
> 
> Cheers, Tim
> 
> -- 
> --
> Tim Gruene
> Institut fuer anorganische Chemie
> Tammannstr. 4
> D-37077 Goettingen
> 
> phone: +49 (0)551 39 22149
> 
> GPG Key ID = A46BEE1A
> 



-- 
--
Tim Gruene
Institut fuer anorganische Chemie
Tammannstr. 4
D-37077 Goettingen

phone: +49 (0)551 39 22149

GPG Key ID = A46BEE1A



signature.asc
Description: Digital signature

Re: [ccp4bb] ctruncate - FP=0?

[Eleanor Dodson]

Are you sure these are real FP=0 or reflections which werent measured 
but have been added for completeness of the h k l list.
The check is whether the SigF is also 0.00 - in that case they are 
genuinely missing..


Eleanor

On 02/09/2011 11:34 PM, Ed Pozharski wrote:

I observe under some conditions that ctruncate sets some reflections
amplitudes to zero.  AFAIU, this should not be happening as even
negative intensities (there are none in this particular dataset) should
produce FP>0 upon truncation.

66 out of ~23000 reflections are zeros after ctruncate is applied.
Nothing obvious comes up upon inspecting the corresponding hkl's, except
that one and only one is always zero (sg is P21212, so these are not
systematic absences).  One curious thing is that the I/sigma for these
reflections is close to the average I/sigma in the highest resolution
range (but it varies and these reflections are in all resolution
ranges).

A bug?

Re: [ccp4bb] Let's talk pseudotranslational symmetry (or maybe it's bad data).

[Eleanor Dodson]

Pseudo translation doesnt seem too much of a problem if the model is OK.
MOLREP uses it within the search, phaser seems not to mind too much for 
a general translation..


I dont know about pathological cases where the crystal is very near but 
not quite C centred, ie the translation is quite special like in this 
case with t = (1/2,0.~1/4)


On 02/09/2011 11:03 PM, Roberts, Sue A - (suer) wrote:

I have used phaser to successfully solve a structure in P2x2x2x that had 
pseudo-translational symmetry.  It was unable to correctly choose the space 
group, but after running phaser in all 8 possible space groups and inspecting 
the best solutions in each the correct solution was clear.  Furthermore, it was 
the only MR program that worked. Was I just lucky?

Sue



On 9 Feb 2011, at 22:27, Phil Jeffrey wrote:


Is there a program that does ?  I was under the impression that they were all 
equally good/bad at this, because any solution that agrees with the PTS has 
quite a high score and any solution that doesn't has a low score, irrespective 
of the correctness of the placement of the molecules.

In one case that ritually defeats me with quite strong pseudo-centering, this 
seems to be true for heavy atom searches also.

Phil Jeffrey
Princeton

On 2/9/11 5:08 PM, Jon Schuermann wrote:

I would NOT use Phaser for MR with PTS present. It doesn't handle it
correctly yet, since the likelihood targets don't account for PTS.
Others may be able to explain it better.




Dr. Sue A. Roberts
Dept. of Chemistry and Biochemistry
University of Arizona
1041 E. Lowell St.,  Tucson, AZ 85721
Phone: 520 621 8171
s...@email.arizona.edu
http://www.biochem.arizona.edu/xray

Re: [ccp4bb] Solubilization refolding percentage efficiency

[Anastassis Perrakis]

How:

1. measure total protein before refolding (eg by absorbance at 280)
2. Purify refolded fraction by gel filtration in FPLC
3. measure total protein after refolding
4. calculate efficiency ...

Why?

I don't see the reason to know the efficiency. If after refolding I  
would get enough in a defined molecular species,
soluble in a non-denaturing buffer, I would be happy. 'Enough' depends  
on what you want to do with it.
Some refolding protocols have efficiencies of just 2-5%, but this is  
enough ...


A.


On Feb 10, 2011, at 10:45, megha goyal wrote:


Hello,

Can anyone let me know how to calculate solubilisation and refolding  
efficiency. We perform solubilisation and then refolding and check  
by HPLC if refolding is completed or not [single peak on HPLC]. how  
do i determine % efficiency of refolding. kindly guide me.


regards,


megha


P please don't print this e-mail unless you really need to
Anastassis (Tassos) Perrakis, Principal Investigator / Staff Member
Department of Biochemistry (B8)
Netherlands Cancer Institute,
Dept. B8, 1066 CX Amsterdam, The Netherlands
Tel: +31 20 512 1951 Fax: +31 20 512 1954 Mobile / SMS: +31 6 28 597791

Re: [ccp4bb] First images of proteins and viruses caught with an X-ray laser

[Colin Nave]

James

All great stuff.

 

Ø  The individual crystals were not twinned (or at least I would be VERY 
surprised if they were)

 

Can this be tested? The spots should show fringes around them corresponding to 
the shape transform of the nano-crystal. If twinned, the intensities of the 
fringes should be modified due to the contrast between the two types of domain 
for the particular reflection. If it worked, one would be equivalent to 
grinding up the crystal using the coherent beam.

 

Probably difficult as would have to be done independently for each crystal. 

 

Colin

 

From: CCP4 bulletin board [mailto:CCP4BB@JISCMAIL.AC.UK] On Behalf Of James 
Holton
Sent: 10 February 2011 01:20
To: CCP4BB@JISCMAIL.AC.UK
Subject: Re: [ccp4bb] First images of proteins and viruses caught with an X-ray 
laser

 

The twin fraction for REFMAC was exactly 0.5.  The individual crystals were not 
twinned (or at least I would be VERY surprised if they were), but they do 
belong to the space group P63, and autoindexing will give you one of the two 
possible axis conventions at random.  So, of the 15445 crystals that were 
merged, there were about ~7700 indexed one way and 7700 indexed the other way.  
It is hard to tell exactly.  Mergeing the data "blindly" (which we did) will 
result in a "twinned data set" where each merged h,k,l has an equal 
contribution from k,h,-l.  This was all discussed in the data processing paper 
PMID: 20389587.  

Obviously, there are a number of things you can think of for resolving this 
indexing ambiguity and removing the "twinning effect", and in fact, this is 
what I thought was really cool about this particular kind of data collection 
and why I encouraged the people doing all the work (the other 87 authors) to go 
ahead with a "twinnable" space group.  On the first pass, you can just take 
advantage of TWIN refinement in REFMAC, but in the future, when the indexing 
improves (probably using post-refinement) it may be possible to de-twin a 
crystal system that actually suffers from "real" twinning.  That is, a "twin 
domain" cannot be smaller than a mosaic block (otherwise the h,k,l and -k,h,-l 
structure factors would add as phased Fs, not |F|^2).  So, since nanocrystals 
are essentially single mosaic blocks (smaller than the coherence length of the 
beam), you could take your twinned crystals and either grind them up or 
re-optimize for smaller crystals (counterintuitive!), and then resolve the twin 
domains "manually".  Sort of like Louis Pasteur and his tartaric acid crystals.

Anyway, I thought that was a cool idea, but like so many other cool things, it 
had to be cut from the Nature paper.  Admittedly, the problem has not actually 
been solved yet.  This is why we used REFMAC in TWIN mode.  

-James Holton
MAD Scientist

On Wed, Feb 9, 2011 at 3:29 PM, Jon Schuermann  wrote:

According to the paper, the data was refined in REFMAC in 'twin mode' which, I 
believe, calculates the R-factor using a  non-conventional R-factor equation 
which usually lower than the conventional R-factor. I believe this is dependent 
on the twin fraction which wasn't mentioned in the paper (or supplementary 
info) unless I missed it.

Jon




-- 
Jonathan P. Schuermann, Ph. D.
Beamline Scientist
NE-CAT, Building 436E
Advanced Photon Source (APS)
Argonne National Laboratory
9700 South Cass Avenue
Argonne, IL 60439
 
email: schue...@anl.gov
Tel: (630) 252-0682
Fax: (630) 252-0687



On 02/09/2011 05:11 PM, James Holton wrote: 

This was "molecular replacement" from 1jb0, so the phases came from the model.  
Probably more properly called "direct refinement" since all we did was a few 
cycles of rigid body.  Personally, I was quite impressed by how good the R 
factors were, all things considered.

-James Holton
MAD Scientist

On Wed, Feb 9, 2011 at 2:56 PM, Bernhard Rupp (Hofkristallrat a.D.) 
 wrote:

Any idea where then phases came from?
BR


-Original Message-
From: CCP4 bulletin board [mailto:CCP4BB@JISCMAIL.AC.UK] On Behalf Of Thomas
Juettemann
Sent: Wednesday, February 09, 2011 12:16 PM
To: CCP4BB@JISCMAIL.AC.UK
Subject: Re: [ccp4bb] First images of proteins and viruses caught with an
X-ray laser

Thank you for clarifying this James. Those details are indeed  often
lost/misinterpreted when the paper is discussed in journal club, so your
comment was especially helpful.

Best wishes,
Thomas

On Wed, Feb 9, 2011 at 20:38, James Holton  wrote:
>
> As one of the people involved (I'm author #74 out of 88 on PMID
> 21293373), I can tell you that about half of the three million
> snapshots were blank, but we wanted to be honest about the number that
> were collected, as well as the "minimum" number that were needed to
> get a useful data set.  The blank images were on purpose, since the
> nanocrystals were diluted so that there would be relatively few
> double-hits.  As many of you know, multiple lattices crash autoindexing
algorithms!
>
> Whether or not a blank image or a failed autoindexing run qualif

Re: [ccp4bb] Let's talk pseudotranslational symmetry (or maybe it's bad data).

[Boaz Shaanan]

Well, I've actually had success using phaser for placing 6 monomers in the a.u. 
P41212 space group BUT only after the first monomer was fully built into the 
initial map (pdb entry 2pan), in other words when the probe was 100% identical 
in sequence and structurally close enough to the target. The PTS peak height in 
the Patterson was ~60% of the origin at 0,0,~0.35). Earlier MR searches in 
phaser, using models that were 25-28% identical, picked up only 3-4 monomers. 

  Cheers,

                Boaz
- Original Message -
From: Phil Jeffrey 
Date: Thursday, February 10, 2011 0:28
Subject: Re: [ccp4bb] Let's talk pseudotranslational symmetry (or maybe it's 
bad data).
To: CCP4BB@JISCMAIL.AC.UK

> Is there a program that does ?  I was under the impression 
> that they 
> were all equally good/bad at this, because any solution that 
> agrees with 
> the PTS has quite a high score and any solution that doesn't has 
> a low 
> score, irrespective of the correctness of the placement of the 
> molecules.
> In one case that ritually defeats me with quite strong pseudo-
> centering, 
> this seems to be true for heavy atom searches also.
> 
> Phil Jeffrey
> Princeton
> 
> On 2/9/11 5:08 PM, Jon Schuermann wrote:
> > I would NOT use Phaser for MR with PTS present. It doesn't 
> handle it
> > correctly yet, since the likelihood targets don't account for PTS.
> > Others may be able to explain it better.
>

Boaz Shaanan, Ph.D.
Dept. of Life Sciences
Ben-Gurion University of the Negev
Beer-Sheva 84105
Israel
Phone: 972-8-647-2220 ; Fax: 646-1710
Skype: boaz.shaanan‎

[ccp4bb] Solubilization refolding percentage efficiency

[megha goyal]

Hello,

Can anyone let me know how to calculate solubilisation and refolding
efficiency. We perform solubilisation and then refolding and check by HPLC
if refolding is completed or not [single peak on HPLC]. how do i determine %
efficiency of refolding. kindly guide me.

regards,


megha

Re: [ccp4bb] Let's talk pseudotranslational symmetry (or maybe it's bad data).

[Randy Read]

Actually, with a non-origin peak height only 1/4 of the origin and a 
translation that does not correspond exactly to a centering operator, Phaser 
has a very good chance of coping with this case.  I think that the space group 
ambiguity of whether the 2-fold parallel to x is a screw axis or a pure 2-fold 
(as Eleanor discussed) is more likely to be the problem.

Regards,

Randy Read

On 9 Feb 2011, at 22:08, Jon Schuermann wrote:

> Just to add to Phil and Eleanor's response...
> 
> I would NOT use Phaser for MR with PTS present. It doesn't handle it 
> correctly yet, since the likelihood targets don't account for PTS. Others may 
> be able to explain it better.
> 
> Its probably not C-centered (as Eleanor mentions) and you should try the 
> other primitive cell options.
> 
> From what I understand about seeing (or not) the strong/weak reflection 
> behavior... it is proportional to the size of your off-origin Patterson peak 
> so it is more difficult to see when the off-origin peak is small (26% in your 
> case). Crystal orientation, I guess, could also be a reason for not 
> visualizing it on the diffraction image, but it still should be apparent in 
> the stats since Xtriage found it.
> 
> If it was me... If your off-origin peak was over 50% of the origin peak 
> height AND after inspecting the Patterson map to see if it lies at z0.25, 
> then you MIGHT be able to solve and refine the structure in C-centered 
> orthorhombic cell. The advantage of this would be that you wouldn't be 
> refining against a large portion of weak reflection data. In your case, 
> however, since your weak reflections are not really that weak you shouldn't 
> have too much of a problem refining.
> 
> Hope this helps.
> 
> Jon
> 
> -- 
> Jonathan P. Schuermann, Ph. D.
> Beamline Scientist
> NE-CAT, Building 436E
> Advanced Photon Source (APS)
> Argonne National Laboratory
> 9700 South Cass Avenue
> Argonne, IL 60439
> 
> email: schue...@anl.gov
> Tel: (630) 252-0682
> Fax: (630) 252-0687
> 
> 
> 
> 
> On 02/08/2011 11:49 AM, Francis E Reyes wrote:
>> Hi all
>> 
>> I have a case of a dataset that indexed, integrated, and scaled well in P 21 
>> 21 21 (55.6410   81.6493  147.1294   90.   90.   90.) . The data 
>> has an Mn(i/sd) of 2.1 at 3.5 A with a Rpim of about 0.398 at the highest 
>> resolution shell (3.49-3.58).
>> 
>> Analysis with phenix.xtriage warns of pseudotranslational symmetry (26% of 
>> origin).
>> 
>> 
>> x  y  zheight   p-value(height)
>> ( 0.500, 0.000, 0.233 ) :   26.344   (2.681e-03)
>> ( 0.000, 0.338, 0.000 ) :5.380   (8.476e-01)
>> 
>> If the observed pseudo translationals are crystallographic
>> the following spacegroups and unit cells are possible:
>> 
>> space groupoperator unit cell of reference setting
>> C 2 2 21 (b-1/4,c-1/4,2*a)   x+1/2, y, z+1/4  (73.64, 55.47, 81.46,  
>> 90.00, 90.00, 90.00)
>> 
>> From what I've read about pseudo c-centering via pseudotranslational 
>> symmetry, the problem exhibits itself with alternating weak and strong 
>> reflections at low resolution, but become consistent at high resolution. 
>> Inspection of the h+k parity groups via truncate does not show this behavior 
>> .
>> 
>> Despite the fact the data was collected at the anomalous peak, I do not 
>> observe any anomalous signal (DelAnom correlation between half-sets is 0.013 
>> for all data).
>> 
>> Using a reasonably complete model (>80%) I searched for two molecules in the 
>> ASU in space group P 21 21 21 and obtained a solution at TFZ=22.1 for two 
>> molecules related solely by a translation.  However the electron density 
>> maps (after rigid body refinement) are not great (or maybe my expectations 
>> are too high). I am encouraged by the fact the density is weak for a region 
>> of the model which should have a different conformation, while strong 
>> density is maintained for the rest of the molecule.
>> 
>> Is this the proper way to approach pseudotranslation (i.e. is there any 
>> reason to believe that the solution obtained by MR is not the correct 
>> solution?).
>> 
>> Is the space group determined? (i.e. does the pseudo c-centering affect 
>> pointless's ability to analyze the systematic absences?).
>> 
>> Is the lack of a pattern of alternating weak/strong reflections normal 
>> (would observing this behavior be dependent on the crystal orientation) ?
>> 
>> any advice would be greatly appreciated! (especially from those who have had 
>> a case like this before)
>> 
>> 
>> F
>> 
>> 
>> -
>> Francis E. Reyes M.Sc.
>> 215 UCB
>> University of Colorado at Boulder
>> 
>> gpg --keyserver pgp.mit.edu --recv-keys 67BA8D5D
>> 
>> 8AE2 F2F4 90F7 9640 28BC  686F 78FD 6669 67BA 8D5D

--
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