Re: [ccp4bb] Anisotropic Scaling

[Gerard Bricogne]

Dear Liang,

 In this case I cannot refrain from mentioning the STARANISO
server at

 http://staraniso.globalphasing.org/

It offers a certain amount of background explanations, with a didactic
aim, that hopefully will help you understand what it does and what its
results provide you with.

 If you like what it does, then you can get all the benefits from
it by letting autoPROC (https://www.globalphasing.com/autoproc/) run
it for you as part of the processing of your raw images.

 Good luck!
 
 
 With best wishes,
 
  Gerard.

--
On Sat, Jan 13, 2018 at 02:52:46PM -0500, Jacqueline Vitali wrote:
> There is an anisotropy server from ucla that I think you can use over the
> internet,  Just google it and try it to see if it helps with your data.  I
> think the data is better afterwards,
> 
> Jackie Vitali
> Cleveland State University
> 
> On Sat, Jan 13, 2018 at 5:53 AM, Zhang Foggy  wrote:
> 
> > Dear Crystallographers,
> >
> > Sorry for the non-ccp4 topic.
> >
> > Recently I collected a set of diffraction data with significant diffraction
> > anisotropy (two directions to 3.0A, while the other to 3.7A). The overall
> > resolution can only be scaled to 3.3A (I/sigma=1.04). I have read some
> > publications that they can scale the data under different directions rather
> > than overall by using HKL3000 (like the table in attached figure), which
> > can describe the anisotropy more accurate. Does anyone can tell me how to
> > scale the data like that?
> >
> > Thank you in advance.
> >
> > Best,
> >
> > Liang
> >

Re: [ccp4bb] Anisotropic Scaling

[Jacqueline Vitali]

There is an anisotropy server from ucla that I think you can use over the
internet,  Just google it and try it to see if it helps with your data.  I
think the data is better afterwards,

Jackie Vitali
Cleveland State University

On Sat, Jan 13, 2018 at 5:53 AM, Zhang Foggy  wrote:

> Dear Crystallographers,
>
> Sorry for the non-ccp4 topic.
>
> Recently I collected a set of diffraction data with significant diffraction
> anisotropy (two directions to 3.0A, while the other to 3.7A). The overall
> resolution can only be scaled to 3.3A (I/sigma=1.04). I have read some
> publications that they can scale the data under different directions rather
> than overall by using HKL3000 (like the table in attached figure), which
> can describe the anisotropy more accurate. Does anyone can tell me how to
> scale the data like that?
>
> Thank you in advance.
>
> Best,
>
> Liang
>

Re: [ccp4bb] "Atomic resolution"

[Robbie Joosten]

Hi Ivan,



I agree that the Hamilton test is quite useful in this context. We added it in 
PDB-REDO quite a while ago 
(http://www.cmbi.ru.nl/pdb_redo/reprints/CCP42011.pdf). Since that paper we 
have learned that at 18 reflections/atom is not enough to consistently pass the 
Hamilton test. We currently only accept anisotropic B-factors without testing 
at 30 reflections/atom.

I would argue that a 1% drop in R-free could still be the result of 
overfitting. After all, you are more than doubling your model parameters.



Cheers,

Robbie



Sent from my Windows 10 phone






From: CCP4 bulletin board  on behalf of Ivan Shabalin 

Sent: Friday, January 12, 2018 6:31:30 PM
To: CCP4BB@JISCMAIL.AC.UK
Subject: Re: [ccp4bb] "Atomic resolution"

Perpetual topic!

The variety of opinions is amazing. Let me add my opinion here.

I like the definition set by Z. Dauter, G. Murshudov, and K. Wilson in
the International tables for Crystallography volume F (2001 edition),
section 18.4.1 "Definition of atomic resolution".

They define the atomic resolution as: when there are sufficient
accurately measured observables to justify the refinement of the ordered
part of the structure with full anisotropic ADPs.

Of course, it has a different meaning than the optical resolution, but
these tend to coincide.

There is no magic number that fits all structures. It depends on the
quality of data, completeness (especially in the highest shell), and
solvent content. 1.2 A is a rather conservative estimate.

To decide whether to use anisotropic ADPs or not, I like utilizing the
Hamilton R-value ratio test.  Basically, once you introduce extra
parameters, the drop of the R-free should be significant in relation to
the increase in the number of parameters.

The concept is very well explained in:

To B or not to B: a question of resolution?
https://www.ncbi.nlm.nih.gov/pubmed/22505267

The test is implemented in HKL3000, and probably, in other software.

If using the Hamilton test is not possible, i substitute the test with
two independent considerations (both are present in Hamilton test):

1) (Number of Unique reflections)/(Number of non-hydrogen atoms)> 18
should put me on rather safe side. This ratio stands for
"data-to-parameter ratio" = 2 or higher. One atom is defined by 9
parameters.

2) Drop of R-free has to be meaningful (significant). My rule of thumb
is that at least 0.5% percent is good. 1% is safer.

Ivan




With best regards,
Ivan Shabalin, Ph.D.
Research Scientist,
Department of Molecular Physiology and Biological Physics,
University of Virginia,
1340 Jefferson Park Avenue, Pinn Hall,Room 4223,
Charlottesville, VA 22908

On 01/11/2018 02:30 PM, Keller, Jacob wrote:
> Dear Crystallographers,
>
> Has there been a consensus as to what is meant by “atomic resolution?”
> Seems like the term is taken by various practitioners to mean different
> things.
>
> A related question: at what resolution are atoms “visible” using only
> the data? I have an empirical feeling that this would be around 1.5 Ang
> Bragg spacings, but on the other hand, one can contour up most maps and
> see individual atom peaks. I would be interested to hear a more rigorous
> way to think about this.
>
> All the best,
>
> Jacob Keller
>
> +
>
> Jacob Pearson Keller
>
> Research Scientist / Looger Lab
>
> HHMI Janelia Research Campus
>
> 19700 Helix Dr, Ashburn, VA 20147
>
> (571)209-4000 x3159
>
> +
>
> The content of this email is confidential and intended for the recipient
> specified in message only. It is strictly forbidden to share any part of
> this message with any third party, without a written consent of the
> sender. If you received this message by mistake, please reply to this
> message and follow with its deletion, so that we can ensure such a
> mistake does not occur in the future.
>

[ccp4bb] Postdoctoral positions available

[Barondeau, David P]

Postdoctoral positions are available immediately in the Barondeau lab at Texas 
A University investigating structure-function properties of the mitochondrial 
iron-sulfur cluster biosynthetic complex.  This project has components of 
structural biology, bioinorganic chemistry, enzymology, and chemical biology.
https://www.chem.tamu.edu/rgroup/barondeau/website/documents/Barondeau-lab-postdoc-PhD-positions.pdf

The project is funded by NIH (renewed in 2017) and focuses on 
structure-function properties of the human Fe-S cluster assembly complex. Fe-S 
clusters are ancient protein cofactors that are required for some of the most 
important reactions in biology. Conserved biosynthetic pathways build and 
distribute these clusters to the hundreds, if not thousands, of proteins that 
require Fe-S clusters for their function. In humans, an Fe-S assembly complex 
located in the mitochondrial matrix is responsible for synthesizing Fe-S 
clusters. Defects in the biogenesis of iron-sulfur clusters are directly 
associated with myopathy, neurodegenerative ataxia and ataxia-susceptibility, 
and contribute to genomic instability, the development of cancer, and aging. 
The structural core of this assembly complex consists of cysteine desulfurase 
(NFS1), eukaryotic-specific LYR protein (ISD11), and acyl carrier protein (ACP) 
subunits and is referred to as the SDA complex. We recently reported crystal 
and electron microscopy structures along with functional properties of the 
mitochondrial cysteine desulfurase (NFS1-ISD11-ACP) complex 
(https://www.ncbi.nlm.nih.gov/pubmed/28634302). This manuscript describes 
lock-and-key interactions between the acyl-chain of ACP and ISD11 along with a 
novel cysteine desulfurase architecture.

Highlights of this study:
http://www.science.tamu.edu/news/story.php?story_ID=1812#.WUlPlmjyuUl
https://www-ssrl.slac.stanford.edu/content/science/highlight/2017-09-30/structure-human-cysteinedesulfurase-complex

Objectives of the project
 1. Apply biophysical methods (X-ray crystallography, SAXS, EM, and mass 
spectrometry based methods) to determine the interactions between the three 
accessory proteins and the core SDA complex that constitute the fully 
functional Fe-S cluster assembly complex.
 2. Elucidate the determinants that drive quaternary structure and activity 
differences between NFS1 and its prokaryotic homolog IscS.
 3. Explore how the composition of the acyl-chain associated with ACP and 
post-translational modifications influence the structure of the assembly 
complex and its ability to synthesize Fe-S clusters.
 4. Determine molecular details of the frataxin activation mechanism for 
Fe-S cluster biosynthesis as a step towards a treatment for Friedreich's ataxia.
 5. Investigate the roles of individual proteins in Fe-S cluster assembly 
and distribution networks using a chemical biology approach coupled to global 
fit kinetic analysis. This strategy takes advantage of an intein-based strategy 
to incorporate fluorophore labels that can be used to report cluster content 
(http://pubs.acs.org/doi/10.1021/ja510998s).

To apply for a Postdoctoral position (initial application deadline February 
1st, 2018), see link below:
https://tamus.wd1.myworkdayjobs.com/en-US/TAMU_External/job/College-Station-TAMU/Postdoctoral-Research-Associate-1_R-000735

Please also feel free to contact me at 
barond...@tamu.edu for more information.


-
David P. Barondeau
Associate Professor
Department of Chemistry
Texas A University
301 Old College Main
College Station, TX 77843
Office: ILSB 1196A
Phone: (979) 458-0735
http://www.chem.tamu.edu/rgroup/barondeau/

Re: [ccp4bb] BUCCANEER label choose

[Eleanor Dodson]

Well - this is a pipeline where cycles of rebuilding are followed by cycles
of refinement.
The Freer is used during refinement - this is always a good idea - but you
need to name the FreeRflag column. I think there is a default

The selection "use map coefficient" and "use PHI/FOM instead of HL
coefficient" are relevent for the Buccaneer cycles.

First  "use PHI/FOM instead of HL coefficient"..  If you are starting from
experimental phases (SAD or MAD)  or parrot modification of these  it is
best to use HL coefficvients , but if you are rebuilding from a MR solution
you may want to start from the 2mFOdFC and PHWT coefficents..


However if you provide the model the pipeline will begin with cycles of
refinement and generate suitable map coeffoicients itself

Eleanor





On 13 January 2018 at 03:28, 张士军 <21620150150...@stu.xmu.edu.cn> wrote:

> Dear all
>
>   I am confusing how to choose or enter the labels when I do density
> modification and autobuilding with ccp4 parrot and buccaneer or ARP/wARP,
> there are three different options like "use Free R-flag ","use map
> coefficient" and "use PHI/FOM instead of HL coefficient" in the figure I
> attached. What they are mean, which one I should choose when I just want to
> do density modification after SAD or MR,and how to do when autobuilding?
> Even though when I choose"use map coefficient", it appeared
> HLA,HLB,HLC,HLD,and when I choose "use PHI/FOM instead of HL coefficient",
> it appeared PHI,FOM, how to enter these labels? Thanks a lot !!!
>
> Best Regards
>
> Shijun
>