Re: [ccp4bb] Free Reflections as Percent and not a Number

[F.Xavier Gomis-Rüth]

As pointed out by a reviewer recently, this has been object of study in 
the past when Gerard K. and Axel B. found that actually 500-700

reflections should suffice to monitor Rfree with sufficient precision.
I guess everybody will agree on that we want to have the best refined 
structure and this entails not setting aside valuable reflections 
unnecessarily.


Kleywegt GJ, Brünger AT. _Checking your imagination: applications of the 
free R value._ Structure. 
1996; 4:897-904.
Brünger AT. _Free R value: cross-validation in crystallography._ 
Methods Enzymol. 
1997;277:366-96


As to the convenience of carrying over a test set to another dataset, 
Eleanor made a suggestion to circumvent this necessity some time ago:

pass your coordinates through pdbset and add some noise before refinement:

pdbset xyzin xx.pdb xyzout yy.pdb <
Dear Crystallographers,

I thought that for reliable values for Rfree, one needs only to satisfy 
counting statistics, and therefore using at most a couple thousand reflections 
should always be sufficient. Almost always, however, some seemingly-arbitrary 
percentage of reflections is used, say 5%. Is there any rationale for using a 
percentage rather than some absolute number like 1000?

All the best,

Jacob

***
Jacob Pearson Keller, PhD
Looger Lab/HHMI Janelia Research Campus
19700 Helix Dr, Ashburn, VA 20147
email: kell...@janelia.hhmi.org
***
.



--

[ccp4bb] PhD position: Virus evolution in immunocompromised patients with chronic norovirus infection

[Grant Hansman]

This PhD project is from the German-Israeli Helmholtz Research School in Cancer 
Biology, a joint PhD training initiative of the German Cancer Research Center 
and the Weizmann Institute of Science in Rehovot, Israel.

The successful candidate will be required to spend 6-12 months during their PhD 
working at the Weizmann Institute (in the laboratory of our collaborator, Prof. 
Joel L. Sussman). For further information about the Research School, please 
refer to the website: https://www.dkfz.de/en/phd-program/GIHRS.html

Immunocompromised patients have an increased risk of norovirus infections and 
the disease can become chronic, lasting weeks to years. Furthermore, it was 
found that immunocompromised patients with chronic norovirus infections could 
generate genetic and antigenic variant viruses, suggesting a mechanism by which 
the virus could escape the host immune responses. The X-ray crystal structure 
of the prototype norovirus particle identified two domains, shell (S) and 
protruding (P) domains. Recently, it was found that the P domain interacts with 
polymorphic human histo-blood group antigens (HBGAs) and this interaction 
appears to be important for virus entry. The goal of this proposal is to better 
understand norovirus evolution in immunocompromised patients with chronic 
norovirus infections. We will concentrate on the RdRp and capsid genes, since 
these are important targets for antiviral development. The project has a number 
of aims:

Determine the full-length genome sequences from norovirus isolates from 
chronically infected patients and describe the genetic changes over time
• Express the variant capsid genes in insect cells and determine the antigenic 
changes among different variant viruses
• Express the variant RdRp gene and capsid P domains in E. coli and determine 
their X-ray crystal structures
• Compare the HBGA binding profiles among the variant P domains using X-ray 
crystallography
• Investigate the enzymatic activities among the variant RdRps
• Develop a virus-tracking system and observe the differences among the 
va-riant VLPs and their attachment/ entry to culture cells

Your profile:
• Experience in standard biochemical and molecular biology techniques as well 
as microbiology/ bacterial culture techniques is required
• Experience in protein purification and some knowledge of X-ray 
crystallogra-phy is desirable
• The successful candidate will be required to spend 6-12 months during their 
PhD working at the Weizmann Institute

Duration:
The position is limited for 3 years.

Contact:
g.hans...@dkfz.de

[ccp4bb] 2nd Reminder - CCP4 Study Weekend 2015 - Early bird registration and Standard Student bursary deadline is 24th November 2014

[Karen McIntyre]

The CCP4 Study Weekend (7 - 9 January 2015)
East Midlands Conference Centre, University of Nottingham

Wednesday 7 - MX User Meeting
Thursday 8 / Friday 9 - CCP4 Study Weekend

"Advances in Experimental Phasing"

We cordially invite you to participate in the 2015 CCP4 Study Weekend at the 
the East Midlands Conference Centre, University of Nottingham. The annual CCP4 
Study Weekend is a chance to shake off the post-New Year torpor, and work hard 
and play hard with your fellow crystallographers. Once again, we have put 
together an exciting scientific programme for the Thursday and Friday, either 
side of the traditional conference dinner. Please also check out the satellite 
meetings which may be of interest. The Study Weekend is a chance to catch up 
with old friends, but is also a chance to meet the CCP4 staff who will be there 
to present and demonstrate the latest software and to answer questions - please 
say hello!

This year, the topic for the Study Weekend is "Advances in Experimental 
Phasing". In keeping with previous CCP4 meetings, the lectures will focus on 
the presentation and discussion of advanced methods and techniques developed 
and used by the leaders in the field.
Scientific Organisers
Thomas Schneider - EMBL Hamburg (Germany)
Airlie McCoy - University of Cambridge (UK)

Further details of the program and the registration are at 
http://www.ccp4.ac.uk/events/CCP4_2015

Please note that we have a limited number of standard student bursaries 
available to any students to help with the cost of attending the Study Weekend. 
The deadline for applicants looking to avail of these bursaries is the 24th of 
November, 2014, so please get your application in before then.
Please note that Early Bird registration closes on 24th November after which 
the cost of registration goes up!
Terms and Conditions apply.  Please read the cancellation policy before 
applying.


-- 
Scanned by iCritical.

[ccp4bb] 2 years Post-doc position near Paris (France)

[Marc Graille]

Post-doctoral position 
“Structural studies of post-transcriptional / translational modification 
enzymes”. 

Location: Palaiseau, Ecole Polytechnique, near Paris (30 minutes by regular 
trains), France

A 2 years post-doctoral position, funded by the ANR (French National Research 
Agency), is available at Laboratory of Biochemistry, Ecole Polytechnique-CNRS 
(France). The successful candidate will work in a young and dynamic 
international team headed by Dr Marc Graille. He/She should have a PhD in 
protein biochemistry/structural biology. Expertises in cloning, protein 
purification and X-ray crystallography are required. Experience in RNA 
production and purification as well as in protein complexes reconstitution will 
be very much appreciated.

The project aims at studying structural aspects of protein-protein and 
protein-RNA complexes involved in the post-transcriptional/translational 
modifications of eukaryotic factors involved in mRNA translation. It will be 
realized in collaboration with groups studying the role of these complexes in 
yeast or human cells.

To address these questions, the successful candidate will use a large panel of 
biochemical (protein expression and purification, enzymology, pull-down, …) and 
biophysical methods (X-ray crystallography, ITC, MALLS, fluorescence, …), which 
are all implemented in the host laboratory (Laboratory of Biochemistry, Ecole 
Polytechnique-CNRS). The lab is also equipped with a state-of-the art 
crystallization platform (Tecan and Mosquito crystallization robots plus an 
automatic system for crystallization plates imaging) and is located 2 km away 
from the French synchrotron SOLEIL, which offers regular access to protein 
crystallography beam lines (Proxima-1 and 2).
Recent publications from Dr Graille’s group related to the project:
Létoquart et al, PNAS, in press.
Figaro et al, Mol. Cell. Biol., 2012
Graille et al, Biochimie, 2012 (review)
Liger et al, Nuc. Acids Res., 2011.
van den Elzen et al; Nat. Struct. Mol. Biol, 2010
More information can be found on our website: 
http://bioc.polytechnique.fr/spip.php?rubrique117&lang=en .

The Laboratory of Biochemistry has a long standing expertise in the study of 
translational apparatus and is located on the multidisciplinary “Ecole 
Polytechnique” campus, one of the most selective and prestigious French “Grande 
Ecole”, which attracts top undergraduate and graduate students worldwide. Ecole 
Polytechnique is located 25 km south from Paris and as regular train 
connections to Paris city centre (30 minutes). The campus offers all the 
equipments to perform various activities (music, theatre, …) or sports: 
swimming pools, 8 tennis courts, several football, volleyball and. basketball 
pitches, climbing walls, … 
More info on the campus can be found at : http://www.polytechnique.edu/en .

Net annual salary: around 24,500€ minimum.
Starting : 2015, depending on availability of selected candidate.
Duration: 24 months.


Applications (including CV, resume describing your relevant skills and a list 
of three references) or information requests should be sent to Marc Graille 
(marc.grai...@polytechnique.edu).
Dr Marc GRAILLE

Directeur de recherche CNRS

Team: Translation and degradation of eukaryotic mRNAs

Team supported by the ATIP-Avenir CNRS program

Laboratoire de Biochimie

ECOLE POLYTECHNIQUE - UMR7654 CNRS

91128 PALAISEAU CEDEX


Phone : +33 (0)1 69 33 48 90 – Fax : +33 (0)1 69 33 49 09


Office 033012B

Email: marc.grai...@polytechnique.edu

http://bioc.polytechnique.fr/spip.php?rubrique117

 
 

Re: [ccp4bb] Free Reflections as Percent and not a Number

[Pavel Afonine]

Hello,

choice of the size of free (or test, whatever you like to call them)
reflections is important for three different purposes:

- estimation of parameters for ML target for refinement;
- map calculation (coefficients m&D in 2mFo-DFc or mFo-DFc map are
calculated using test reflections);
- validation (calculation Rfree).

It is important that free reflections are evenly distributed across the
whole resolution range, and each sufficiently thin resolution bin contains
at least 50 test reflections so that the estimation of ML parameters is
robust and reliable. "Sufficiently thin resolution bin" is such that ML
parameters can be assumed constants in it.

Smaller test sets will result in less stable refinements (refinement
outcome will strongly depend on the choice of test set).

Larger test sets will damage map quality (unless all reflections are used
in map calculation).

Size of free set needs to be sufficiently large so that Rfree is
statistically meaningful.

Nothing new is said above, it's all documented in the literature!

Pavel


On Thu, Nov 20, 2014 at 2:43 PM, Keller, Jacob 
wrote:

> Dear Crystallographers,
>
> I thought that for reliable values for Rfree, one needs only to satisfy
> counting statistics, and therefore using at most a couple thousand
> reflections should always be sufficient. Almost always, however, some
> seemingly-arbitrary percentage of reflections is used, say 5%. Is there any
> rationale for using a percentage rather than some absolute number like 1000?
>
> All the best,
>
> Jacob

Re: [ccp4bb] Free Reflections as Percent and not a Number

[Keller, Jacob]

Agree with all of this—but how does it reflect on the original question of 
whether to use a percent or an absolute number?

JPK

From: Pavel Afonine [mailto:pafon...@gmail.com]
Sent: Friday, November 21, 2014 11:02 AM
To: Keller, Jacob
Cc: CCP4BB@JISCMAIL.AC.UK
Subject: Re: [ccp4bb] Free Reflections as Percent and not a Number

Hello,

choice of the size of free (or test, whatever you like to call them) 
reflections is important for three different purposes:

- estimation of parameters for ML target for refinement;
- map calculation (coefficients m&D in 2mFo-DFc or mFo-DFc map are calculated 
using test reflections);
- validation (calculation Rfree).

It is important that free reflections are evenly distributed across the whole 
resolution range, and each sufficiently thin resolution bin contains at least 
50 test reflections so that the estimation of ML parameters is robust and 
reliable. "Sufficiently thin resolution bin" is such that ML parameters can be 
assumed constants in it.

Smaller test sets will result in less stable refinements (refinement outcome 
will strongly depend on the choice of test set).

Larger test sets will damage map quality (unless all reflections are used in 
map calculation).

Size of free set needs to be sufficiently large so that Rfree is statistically 
meaningful.

Nothing new is said above, it's all documented in the literature!

Pavel


On Thu, Nov 20, 2014 at 2:43 PM, Keller, Jacob 
mailto:kell...@janelia.hhmi.org>> wrote:
Dear Crystallographers,

I thought that for reliable values for Rfree, one needs only to satisfy 
counting statistics, and therefore using at most a couple thousand reflections 
should always be sufficient. Almost always, however, some seemingly-arbitrary 
percentage of reflections is used, say 5%. Is there any rationale for using a 
percentage rather than some absolute number like 1000?

All the best,

Jacob

Re: [ccp4bb] Free Reflections as Percent and not a Number

[Pavel Afonine]

Oh I see, I though the answer follows from that. Fraction is better (or may
be fraction with a cap). Hardwiring a number may not always work. For small
crystals or small data sets or incomplete datasets say 1000 reflections may
mean 50% of the dataset.

All the best,
Pavel

On Fri, Nov 21, 2014 at 8:09 AM, Keller, Jacob 
wrote:

>  Agree with all of this—but how does it reflect on the original question
> of whether to use a percent or an absolute number?
>
>
>
> JPK
>
>
>
> *From:* Pavel Afonine [mailto:pafon...@gmail.com]
> *Sent:* Friday, November 21, 2014 11:02 AM
> *To:* Keller, Jacob
> *Cc:* CCP4BB@JISCMAIL.AC.UK
> *Subject:* Re: [ccp4bb] Free Reflections as Percent and not a Number
>
>
>
> Hello,
>
>
>
> choice of the size of free (or test, whatever you like to call them)
> reflections is important for three different purposes:
>
>
>
> - estimation of parameters for ML target for refinement;
>
> - map calculation (coefficients m&D in 2mFo-DFc or mFo-DFc map are
> calculated using test reflections);
>
> - validation (calculation Rfree).
>
>
>
> It is important that free reflections are evenly distributed across the
> whole resolution range, and each sufficiently thin resolution bin contains
> at least 50 test reflections so that the estimation of ML parameters is
> robust and reliable. "Sufficiently thin resolution bin" is such that ML
> parameters can be assumed constants in it.
>
>
>
> Smaller test sets will result in less stable refinements (refinement
> outcome will strongly depend on the choice of test set).
>
>
>
> Larger test sets will damage map quality (unless all reflections are used
> in map calculation).
>
>
>
> Size of free set needs to be sufficiently large so that Rfree is
> statistically meaningful.
>
>
>
> Nothing new is said above, it's all documented in the literature!
>
>
>
> Pavel
>
>
>
>
>
> On Thu, Nov 20, 2014 at 2:43 PM, Keller, Jacob 
> wrote:
>
> Dear Crystallographers,
>
> I thought that for reliable values for Rfree, one needs only to satisfy
> counting statistics, and therefore using at most a couple thousand
> reflections should always be sufficient. Almost always, however, some
> seemingly-arbitrary percentage of reflections is used, say 5%. Is there any
> rationale for using a percentage rather than some absolute number like 1000?
>
> All the best,
>
> Jacob
>

Re: [ccp4bb] Free Reflections as Percent and not a Number

[Keller, Jacob]

Right about the 1000 in that case, but also Rfree with 5% would be 
statistically poor. I guess one would be stuck in that case.

JPK

From: Pavel Afonine [mailto:pafon...@gmail.com]
Sent: Friday, November 21, 2014 11:16 AM
To: Keller, Jacob
Cc: CCP4BB@JISCMAIL.AC.UK
Subject: Re: [ccp4bb] Free Reflections as Percent and not a Number

Oh I see, I though the answer follows from that. Fraction is better (or may be 
fraction with a cap). Hardwiring a number may not always work. For small 
crystals or small data sets or incomplete datasets say 1000 reflections may 
mean 50% of the dataset.

All the best,
Pavel

On Fri, Nov 21, 2014 at 8:09 AM, Keller, Jacob 
mailto:kell...@janelia.hhmi.org>> wrote:
Agree with all of this—but how does it reflect on the original question of 
whether to use a percent or an absolute number?

JPK

From: Pavel Afonine [mailto:pafon...@gmail.com]
Sent: Friday, November 21, 2014 11:02 AM
To: Keller, Jacob
Cc: CCP4BB@JISCMAIL.AC.UK
Subject: Re: [ccp4bb] Free Reflections as Percent and not a Number

Hello,

choice of the size of free (or test, whatever you like to call them) 
reflections is important for three different purposes:

- estimation of parameters for ML target for refinement;
- map calculation (coefficients m&D in 2mFo-DFc or mFo-DFc map are calculated 
using test reflections);
- validation (calculation Rfree).

It is important that free reflections are evenly distributed across the whole 
resolution range, and each sufficiently thin resolution bin contains at least 
50 test reflections so that the estimation of ML parameters is robust and 
reliable. "Sufficiently thin resolution bin" is such that ML parameters can be 
assumed constants in it.

Smaller test sets will result in less stable refinements (refinement outcome 
will strongly depend on the choice of test set).

Larger test sets will damage map quality (unless all reflections are used in 
map calculation).

Size of free set needs to be sufficiently large so that Rfree is statistically 
meaningful.

Nothing new is said above, it's all documented in the literature!

Pavel


On Thu, Nov 20, 2014 at 2:43 PM, Keller, Jacob 
mailto:kell...@janelia.hhmi.org>> wrote:
Dear Crystallographers,

I thought that for reliable values for Rfree, one needs only to satisfy 
counting statistics, and therefore using at most a couple thousand reflections 
should always be sufficient. Almost always, however, some seemingly-arbitrary 
percentage of reflections is used, say 5%. Is there any rationale for using a 
percentage rather than some absolute number like 1000?

All the best,

Jacob

Re: [ccp4bb] Free Reflections as Percent and not a Number

[Tim Gruene]

Hi Jacob,

If you have a sufficient number of unique reflections, use percentage
because it is more convenient. If not, at least make sure the below
criteria are met. If that's not possible, don't use maximum likelihood
for refinement ;-)

Cheers,
Tim

On 11/21/2014 05:09 PM, Keller, Jacob wrote:
> Agree with all of this—but how does it reflect on the original question of 
> whether to use a percent or an absolute number?
> 
> JPK
> 
> From: Pavel Afonine [mailto:pafon...@gmail.com]
> Sent: Friday, November 21, 2014 11:02 AM
> To: Keller, Jacob
> Cc: CCP4BB@JISCMAIL.AC.UK
> Subject: Re: [ccp4bb] Free Reflections as Percent and not a Number
> 
> Hello,
> 
> choice of the size of free (or test, whatever you like to call them) 
> reflections is important for three different purposes:
> 
> - estimation of parameters for ML target for refinement;
> - map calculation (coefficients m&D in 2mFo-DFc or mFo-DFc map are calculated 
> using test reflections);
> - validation (calculation Rfree).
> 
> It is important that free reflections are evenly distributed across the whole 
> resolution range, and each sufficiently thin resolution bin contains at least 
> 50 test reflections so that the estimation of ML parameters is robust and 
> reliable. "Sufficiently thin resolution bin" is such that ML parameters can 
> be assumed constants in it.
> 
> Smaller test sets will result in less stable refinements (refinement outcome 
> will strongly depend on the choice of test set).
> 
> Larger test sets will damage map quality (unless all reflections are used in 
> map calculation).
> 
> Size of free set needs to be sufficiently large so that Rfree is 
> statistically meaningful.
> 
> Nothing new is said above, it's all documented in the literature!
> 
> Pavel
> 
> 
> On Thu, Nov 20, 2014 at 2:43 PM, Keller, Jacob 
> mailto:kell...@janelia.hhmi.org>> wrote:
> Dear Crystallographers,
> 
> I thought that for reliable values for Rfree, one needs only to satisfy 
> counting statistics, and therefore using at most a couple thousand 
> reflections should always be sufficient. Almost always, however, some 
> seemingly-arbitrary percentage of reflections is used, say 5%. Is there any 
> rationale for using a percentage rather than some absolute number like 1000?
> 
> All the best,
> 
> Jacob
> 

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

GPG Key ID = A46BEE1A



signature.asc
Description: OpenPGP digital signature

[ccp4bb] Formulation of my own mother liquor

[amro selem]

Dear All,first i wish you nice weekend, then i wanna ask about formulation of 
my own mother liquor, i want to optimize a condition containing 0.2m MgCl2, 
0.1m Tris , PH 8; and 20% PEG6K as pricipitant. (PACT)the question is , should 
i make a stock solution from every ingredient by dissolving it in miliQ water. 
then mix all stuff  together  and add water till final concentration with out 
adjusting final pH. OR dissolve the PEG in pH adjusted buffer and then mix all 
ingredients , so the final pH will be 8.any suggest any idea or any one knows 
what the company do. CHEERSAmr

[ccp4bb] postdoctoral position available

[Collins, Edward J]

POSTDOCTORAL FELLOW POSITION University of North Carolina DEPARTMENT OF 
MICROBIOLOGY AND IMMUNOLOGY

We are seeking a motivated PhD crystallographer to drive two projects dealing 
with host defense against pathogens.  The first project involves peptide 
binding to MHC and is a highly productive collaboration between a computational 
chemist and our lab involving design of HIV therapeutics.  The second project 
is the design and production of therapeutic bi-specific antibodies directed 
against peptide/MHC complexes.  We have already designed and are currently 
testing an antibody directed towards HER2/neu overexpressing tumors.   The 
fellow will be responsible for the crystallography associated with each project 
and can drive the therapeutic antibody project as far as they wish to go.

The position is open immediately and is available for up to four years with 
yearly reappointments.

Biophysicists without direct crystallographic experience will be considered.

Applications will be administered through the department at: 
https://unc.peopleadmin.com/postings/63270
Please include a C.V., cover letter and list of three references.

The cover letter should be directed specifically to  Dr. Edward J Collins and 
specifying that you are applying for this set of projects.

Since the applications on the website go to our department, it would be wise to 
send an email to Dr. Collins once the application is complete.

Re: [ccp4bb] Free Reflections as Percent and not a Number

[Pavel Afonine]

Yes, that's true. In such corner cases phenix.refine switches to LS, for
example. Alternatively you can generate many (say 100) different tests sets
and run 100 refinements. This will give you an ensemble of slightly
different structures with slightly different Rwork and Rfree. You can then
derive an uncertainty from this due to the choice of test set.

Paper by Praznikar & Turk in upcoming Acta D will provide a methodological
solution to this. Yet, it may be a long way before it appears as routine in
refinement programs.

Pavel


On Fri, Nov 21, 2014 at 8:22 AM, Keller, Jacob 
wrote:

>  Right about the 1000 in that case, but also Rfree with 5% would be
> statistically poor. I guess one would be stuck in that case.
>
>
>
> JPK
>
>
>
> *From:* Pavel Afonine [mailto:pafon...@gmail.com]
> *Sent:* Friday, November 21, 2014 11:16 AM
>
> *To:* Keller, Jacob
> *Cc:* CCP4BB@JISCMAIL.AC.UK
> *Subject:* Re: [ccp4bb] Free Reflections as Percent and not a Number
>
>
>
> Oh I see, I though the answer follows from that. Fraction is better (or
> may be fraction with a cap). Hardwiring a number may not always work. For
> small crystals or small data sets or incomplete datasets say 1000
> reflections may mean 50% of the dataset.
>
>
>
> All the best,
>
> Pavel
>
>
>
> On Fri, Nov 21, 2014 at 8:09 AM, Keller, Jacob 
> wrote:
>
> Agree with all of this—but how does it reflect on the original question of
> whether to use a percent or an absolute number?
>
>
>
> JPK
>
>
>
> *From:* Pavel Afonine [mailto:pafon...@gmail.com]
> *Sent:* Friday, November 21, 2014 11:02 AM
> *To:* Keller, Jacob
> *Cc:* CCP4BB@JISCMAIL.AC.UK
> *Subject:* Re: [ccp4bb] Free Reflections as Percent and not a Number
>
>
>
> Hello,
>
>
>
> choice of the size of free (or test, whatever you like to call them)
> reflections is important for three different purposes:
>
>
>
> - estimation of parameters for ML target for refinement;
>
> - map calculation (coefficients m&D in 2mFo-DFc or mFo-DFc map are
> calculated using test reflections);
>
> - validation (calculation Rfree).
>
>
>
> It is important that free reflections are evenly distributed across the
> whole resolution range, and each sufficiently thin resolution bin contains
> at least 50 test reflections so that the estimation of ML parameters is
> robust and reliable. "Sufficiently thin resolution bin" is such that ML
> parameters can be assumed constants in it.
>
>
>
> Smaller test sets will result in less stable refinements (refinement
> outcome will strongly depend on the choice of test set).
>
>
>
> Larger test sets will damage map quality (unless all reflections are used
> in map calculation).
>
>
>
> Size of free set needs to be sufficiently large so that Rfree is
> statistically meaningful.
>
>
>
> Nothing new is said above, it's all documented in the literature!
>
>
>
> Pavel
>
>
>
>
>
> On Thu, Nov 20, 2014 at 2:43 PM, Keller, Jacob 
> wrote:
>
> Dear Crystallographers,
>
> I thought that for reliable values for Rfree, one needs only to satisfy
> counting statistics, and therefore using at most a couple thousand
> reflections should always be sufficient. Almost always, however, some
> seemingly-arbitrary percentage of reflections is used, say 5%. Is there any
> rationale for using a percentage rather than some absolute number like 1000?
>
> All the best,
>
> Jacob
>
>
>

Re: [ccp4bb] Free Reflections as Percent and not a Number

[Dale Tronrud]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1



On 11/21/2014 12:35 AM, "F.Xavier Gomis-Rüth" wrote:
> 
> 
> As to the convenience of carrying over a test set to another
> dataset, Eleanor made a suggestion to circumvent this necessity
> some time ago: pass your coordinates through pdbset and add some
> noise before refinement:
> 
> pdbset xyzin xx.pdb xyzout yy.pdb < 

   I've heard this "debiasing" procedure proposed before, but I've
never seen a proper test showing that it works.  I'm concerned that
this will not erase the influence of low resolution reflections that
were in the old working set but are now in the new test set.  While
adding 0.4 A gaussian noise to a model would cause large changes to
the 2 A structure factors I doubt it would do much to those at 10 A.

   It seems to me that one would have to have random, but correlated,
shifts in atomic parameters to affect the low resolution data - waves
of displacements, sometimes to the left and other times to the right.
 You would need, of course, a superposition of such waves that span
all the scales of resolution in the data set.

   Has anyone looked at the pdbset jiggling results and shown that the
low resolution data are scrambled?

Dale Tronrud

> Xavier
> 
> On 20/11/14 11:43 PM, Keller, Jacob wrote:
>> Dear Crystallographers,
>> 
>> I thought that for reliable values for Rfree, one needs only to
>> satisfy counting statistics, and therefore using at most a couple
>> thousand reflections should always be sufficient. Almost always,
>> however, some seemingly-arbitrary percentage of reflections is
>> used, say 5%. Is there any rationale for using a percentage
>> rather than some absolute number like 1000?
>> 
>> All the best,
>> 
>> Jacob
>> 
>> *** Jacob Pearson Keller,
>> PhD Looger Lab/HHMI Janelia Research Campus 19700 Helix Dr,
>> Ashburn, VA 20147 email: kell...@janelia.hhmi.org 
>> *** .
>> 
> 
> --
-BEGIN PGP SIGNATURE-
Version: GnuPG v2.0.22 (MingW32)

iEYEARECAAYFAlRviu4ACgkQU5C0gGfAG12TMwCfTT0Q4yfCCOxJlRXtsCXmmp1n
9lEAn2Ir57+Y16fh02VcsvDxwu6KYRGK
=68gK
-END PGP SIGNATURE-

Re: [ccp4bb] Formulation of my own mother liquor

[Janet Newman]

Generally one will make this up from stock solutions:

Make up
1 M Tris chloride pH 8
1 M magnesium chloride
50 % (weight/volume) PEG 6K

Then mix these together in the right ratios

(for 1 mL)

100 uL 1 M tris solution
200 uL 1 M MgCl2 solution
400 uL 50% PEG solution
300 uL water.




Janet Newman
Principal Scientist / Director, Collaborative Crystallisation Centre
CSIRO Material Science and Engineering
343 Royal Parade
Parkville.  VIC. 3052
Australia
Tel +613 9662 7326
Email janet.new...@csiro.au

From: CCP4 bulletin board [CCP4BB@JISCMAIL.AC.UK] on behalf of amro selem 
[03391c09f749-dmarc-requ...@jiscmail.ac.uk]
Sent: 22 November 2014 03:48
To: CCP4BB@JISCMAIL.AC.UK
Subject: [ccp4bb] Formulation of my own mother liquor

Dear All,
first i wish you nice weekend, then i wanna ask about formulation of my own 
mother liquor, i want to optimize a condition containing
 0.2m MgCl2, 0.1m Tris , PH 8; and 20% PEG6K as pricipitant. (PACT)
the question is , should i make a stock solution from every ingredient by 
dissolving it in miliQ water. then mix all stuff  together  and add water till 
final concentration with out adjusting final pH. OR dissolve the PEG in pH 
adjusted buffer and then mix all ingredients , so the final pH will be 8.
any suggest any idea or any one knows what the company do.
CHEERS
Amr

Re: [ccp4bb] Formulation of my own mother liquor

[David Schuller]

The most important thing is that you do it the same way every time.
But you might also put some thought into flexibility, in case you need 
to explore cryoprotectants, etc.


On 11/21/14 11:48, amro selem wrote:

Dear All,
first i wish you nice weekend, then i wanna ask about formulation of 
my own mother liquor, i want to optimize a condition containing

 0.2m MgCl2, 0.1m Tris , PH 8; and 20% PEG6K as pricipitant. (PACT)
the question is , should i make a stock solution from every ingredient 
by dissolving it in miliQ water. then mix all stuff  together  and add 
water till final concentration with out adjusting final pH. OR 
dissolve the PEG in pH adjusted buffer and then mix all ingredients , 
so the final pH will be 8.

any suggest any idea or any one knows what the company do.
CHEERS
Amr




--
===
All Things Serve the Beam
===
   David J. Schuller
   modern man in a post-modern world
   MacCHESS, Cornell University
   schul...@cornell.edu

[ccp4bb] Restraints after refinement

[jeorgemarley thomas]

Dear all,

I am working on my molecule, but after running

Refmac5 restrained refinement the bonding distances and angles are broken.

I used Phenix to create the ligand and used Coot to place the ligand
into the correct location in the molecule, merged the ligand and
molecule coordinates, and saved the file. Then I went to run
restrained refinement on
the merged coordinates using the cif file provided from the phenix to
the refmac library,the rest of the settings were set at the defaults.
During fitting the model in the 2Fo-Fc map of the ligand there were no
issue, but it fitted well and after refinement of the same, the bond
breaks and
distorted as observed in coot. The resolution is 1.6 Ang.
phosphate linkage bond is broken. What could be the best possible
reason for this and what can be solution for this ? Could you please
suggest what should I do for a complete model for the ligand ? I would
like to run a round of restrained refinement without breaking any
geometric restraints that should be imposed on my ligand.

Thank you for your help in advance,


Jeorge

Re: [ccp4bb] Restraints after refinement

[CHAVAS Leonard]

Dear Jeorge

just as a curiosity:
1- do you see a continuous electron density for the ligand, before and after 
performing the refinement?
2- do you have only one copy of the molecule(s), and if not, do you see the 
same behaviour for all the ligands?
3- what are the B-factor values for the ligand? Any hint that the phosphate 
could be very flexible and actually missing?
4- do you see any red (/ green) FoFc densities where the phosphate should (/ 
not) be? If so, do you see anything else by using 'less data', i.e. data that 
would be less prone to radiation damage?
5- what is the resolution of your data?
6- is it a known ligand? If you do have the same ligand, even only once, 
present in the PDB, why don't you take the coordinates and libraries from this 
already refined structure and try to fit it in your coordinates?
7- do you have chemical data showing that the ligand can stay for very long 
times in your protein without being shortened?
8- do you know how pure is your ligand solution? i.e. is your ligand stable at 
room temperature (or whatever temperature you did the soaking / complex 
formation).
9- similarly, if you did co-crystallisation instead of soaking, it could be 
that your protein somehow cleaved the molecule. It will all depend on how 
stable is the ligand in high concentrations of the protein, pH dependency and 
more.

Those are not real answers to your original question, just points of 
curiosity...

Cheers, Leo

> On Nov 22, 2014, at 4:26 AM, jeorgemarley thomas  
> wrote:
> 
> Dear all, 
> 
> I am working on my molecule, but after running
> Refmac5 restrained refinement the bonding distances and angles are broken. 
> I used Phenix to create the ligand and used Coot to place the ligand into the 
> correct location in the molecule, merged the ligand and molecule coordinates, 
> and saved the file. Then I went to run restrained refinement on
> the merged coordinates using the cif file provided from the phenix to the 
> refmac library,the rest of the settings were set at the defaults. During 
> fitting the model in the 2Fo-Fc map of the ligand there were no issue, but it 
> fitted well and after refinement of the same, the bond breaks and 
> distorted as observed in coot. The resolution is 1.6 Ang. 
> phosphate linkage bond is broken. What could be the best possible reason for 
> this and what can be solution for this ? Could you please suggest what should 
> I do for a complete model for the ligand ? I would like to run a round of 
> restrained refinement without breaking any geometric restraints that should 
> be imposed on my ligand.   
> 
> Thank you for your help in advance,
> 
> Jeorge
>