[ccp4bb] AW: [ccp4bb] Problem with MolRep

[Herman . Schreuder]

Dear Madhurima,

Small protein structures can be very difficult to solve by MR due to an 
unfavorable ratio between inter- and intra-molecular vectors in the pattersons. 
I am currently also struggling with some data sets myself.

However, there are things you could (and should) try:

1) If your search protein forms an oligomer, e.g. dimer, trimer etc., 
generate this oligomer and use this as a search model. Having a bigger search 
model will significantly increase your signal to noise ratio.

2) Try as many different MR programs as you can lay your hand on. Sometimes 
one succeeds, where others fail. I would at least try phaser.

3) Do not run the program(s) using the default parameters, but carefully 
read the manual and adjust the parameters based on the size and shape of your 
search protein. If your protein contains cysteines and/or methionines, you 
could look if there is any anomalous signal in your home source data. Getting 
higher resolution synchrotron data may not hurt as well.

Good luck!
Herman


Von: CCP4 bulletin board [mailto:CCP4BB@JISCMAIL.AC.UK] Im Auftrag von 
Madhurima Roy
Gesendet: Donnerstag, 9. Februar 2017 05:18
An: CCP4BB@JISCMAIL.AC.UK
Betreff: [ccp4bb] Problem with MolRep

Hi all,
I have a small protein which is of 6 kDa including six histidine tag. The 
protein crystallized in the conditions given bellow :
a)  0.1M Sodium acetate trihydrate pH 4.6, 3M NaCl.
b)  0.1 M Bis-tris pH 5.5, 2M Ammonium sulphate.
The crystals are plate shaped in both conditions.We collected diffraction data 
at our home source using Rigaku RaxisIV and processed the data using XDS.

In spite of good homology, the protein structure cannot be solved by MolRep , 
the contrast is very low approx 1.65. The PDB Blast result is given below:

38.1(total score), 70%(query coverage) 1e-05 (E-value) and 50%(Identity) .
The screen shot of the statistics showing R factor and quality of fit in  
CORRECT.LP file is enclosed below.


Kindly help.

Thanks in advance.
Madhurima

Re: [ccp4bb] Composit omit map vs. ligand

[Petr Kolenko]

Dear colleagues,

First of all, I would like to thank to all who responded! I have 
selected some answers and my reactions to sum up:


1) Kay Diedrichs - try a Polder map: clear density for the ligand at 3 
sigma appeared.
2) JR Helliwell - build first five atoms and check the density now (read 
Minichino et al Acta Cryst D 2002): clear difference density for another 
eight atoms with correct geometry appeared. I did not do this 
previously, because these five atoms are half of a compact ring. I was 
sure that  I must see the WHOLE ring in case of the right observation.


These two checks together with fine refinement of the ligand with 
reduced occupancy convinced me to use the dataset. Initially, I thought 
the dataset is useless. Take-home message for me:


- Do not give up too early and try more calculation procedures with 
every dataset you have.

- Go through you old non-complex data and try new procedures.

There are some other calculations to perform (e.g. Fo,soak - Fo,native 
map calculation). And although I am not a big fan of huge data storage, 
I will deposit the diffraction images together with the coordinates 
especially for this structure!


Once again, thanks for your help!
Petr


Dne 08.02.2017 v 9:05 Petr Kolenko napsal(a):

Dear colleagues,

we have a dataset with potential enzyme:ligand complex at 2.2 AA 
resolution. The ligand is very good substrate for the enzyme, we used 
soaking. We do not see the ligand in the regular difference electron 
density, only five out of twenty atoms. However, the ligand placed at 
the active site (model used from structure of a mutant variant) is 
refined well, giving no negative peaks in difference electron density 
map and nice observed electron density. I have calculated composit 
omit map with annealing in Phenix (input model did not contain the 
ligand) and the electron density for the ligand is there.


I have my own opinion, but we are desperate to obtain such data (more 
than 40 crystals already tested). My question is, would this be proof 
of presence of the ligand with reduced occupancy? Will this map 
convince the reviewers? Is there any other way to validate presence of 
the ligand?


Best regards,
Petr

[ccp4bb] PhD positions ITN ES-Cat

[A.M.W.H. Thunnissen]

PhD positions are open in the field of structure-based protein engineering for 
biocatalysis, as part of a Marie Curie Innovative Training Network.

Two 4-year positions as a researcher that should lead to a PhD are open at the 
University of Groningen, the Netherlands. Part of the Marie Curie Innovative 
Training Network ES-Cat, aimed at directed protein evolution for synthetic 
biology and biocatalysis. The University of Groningen subprojects (supervised 
by prof Dick Janssen, Biotransformation and Biocatalysis group) focus on the 
incorporation of computational methods in enzyme engineering (e.g., Wijma et 
al. (2015) Angew Chem 127: 3797–801), and the use of protein X-ray 
crystallography for validation and further improvement of 
in-silico/experimental work flows. Enzymes under investigation are epoxide 
conversion enzymes and related α/β-hydrolase fold enzymes of industrial 
relevance.

Starting date: March-April 2017

Links with more info:   http://www.bioc.cam.ac.uk/hollfelder/Research/es-cat 
  
http://www.rug.nl/research/biotransformation-biocatalysis 


Important qualifications: MsC degree in (bio)chemistry, bioorganic chemistry, 
structural biology, biophysical chemistry, applied biocatalysis, or similar. 
Experience with structural biology, molecular modeling and simulations, enzyme 
isolation and characterization, biocatalytic conversions, and relevant 
analytical methods (GC, HPLC, MS, NMR..).

Personal traits: Organized, reliable, careful and ambitious. Independence and 
perseverance. Team worker, open to collaboration with other researchers. Good 
oral and written command of English.

For further information: contact Andy Thunnissen at a.m.w.h.thunnis...@rug.nl 


-- 

Dr. A.M.W.H. Thunnissen | Protein Crystallography
Laboratory of Biophysical Chemistry, GBB Institute, University of Groningen
Nijenborgh 7, 9747 AG Groningen, the Netherlands
Phone +31.50.3634380 | Fax +31.50.3634800 | E-mail a.m.w.h.thunnis...@rug.nl

Re: [ccp4bb] AW: [ccp4bb] Problem with MolRep

[Eleanor Dodson]

Check solvent % - as suggested low solvent content means tight packing
means poor signal.

Check matthews coefficient ? Do you expect several molecules?
As suggested - does the model form an oligamer? ( You need to be careful
with these - sometimes you generate thm using both crystal and
non-crystallographic symmetry..

Check possible alternate SGs.

Try pruning your model - look at it in CCP4MG and see if there are extruded
loops..
And good luck!

Eleanor

PS - by the way - you are cutting your data very savagely..
CC1/2 > 0.5 acceptable.

On 9 February 2017 at 08:20,  wrote:

> Dear Madhurima,
>
>
>
> Small protein structures can be very difficult to solve by MR due to an
> unfavorable ratio between inter- and intra-molecular vectors in the
> pattersons. I am currently also struggling with some data sets myself.
>
>
>
> However, there are things you could (and should) try:
>
> 1) If your search protein forms an oligomer, e.g. dimer, trimer etc.,
> generate this oligomer and use this as a search model. Having a bigger
> search model will significantly increase your signal to noise ratio.
>
> 2) Try as many different MR programs as you can lay your hand on.
> Sometimes one succeeds, where others fail. I would at least try phaser.
>
> 3) Do not run the program(s) using the default parameters, but
> carefully read the manual and adjust the parameters based on the size and
> shape of your search protein. If your protein contains cysteines and/or
> methionines, you could look if there is any anomalous signal in your home
> source data. Getting higher resolution synchrotron data may not hurt as
> well.
>
>
>
> Good luck!
>
> Herman
>
>
>
>
>
> *Von:* CCP4 bulletin board [mailto:CCP4BB@JISCMAIL.AC.UK] *Im Auftrag von
> *Madhurima Roy
> *Gesendet:* Donnerstag, 9. Februar 2017 05:18
> *An:* CCP4BB@JISCMAIL.AC.UK
> *Betreff:* [ccp4bb] Problem with MolRep
>
>
>
> Hi all,
>
> I have a small protein which is of 6 kDa including six histidine tag. The
> protein crystallized in the conditions given bellow :
>
> a)  0.1M Sodium acetate trihydrate pH 4.6, 3M NaCl.
>
> b)  0.1 M Bis-tris pH 5.5, 2M Ammonium sulphate.
>
> The crystals are plate shaped in both conditions.We collected diffraction
> data at our home source using Rigaku RaxisIV and processed the data using
> XDS.
>
>
> In spite of good homology, the protein structure cannot be solved by
> MolRep , the contrast is very low approx 1.65. The PDB Blast result is
> given below:
>
>
> 38.1(total score), 70%(query coverage) 1e-05 (E-value) and 50%(Identity) .
>
> The screen shot of the statistics showing R factor and quality of fit in
> CORRECT.LP file is enclosed below.
>
>
>
> Kindly help.
>
> Thanks in advance.
>
> Madhurima
>
>
>
>
>
>
>
>
>
>
>

[ccp4bb] The paper describing wwPDB OneDep system is now available.

[Jasmine Young]

The paper 
 
describing wwPDB OneDep system  is now 
available. The wwPDB has deployed a unified system for deposition, 
biocuration, and validation of macromolecular structures globally across 
all wwPDB, EMDB, and BMRB deposition sites to meet the evolving 
requirements of the scientific community to archive structural data over 
the coming decades.


The OneDep system provides a user-friendly deposition interface and 
improved structure validation  with the 
benefit of recommendations from expert task forces 
 representing the 
respective methodological communities. The processing efficiency in 
biocuration is improved as OneDep supports a more automated workflow .


As Milka Kostic, the Senior Editor at */Structure/* and */Cell Chemical 
Biology/* described, OneDep is a step in the right direction 
 
and offers a single point of entry into the atomic coordinate deposition 
process, as well as improving processes of structure validation and data 
biocuration.



--
Regards,

Jasmine

===
Jasmine Young, Ph.D.
Biocuration Team Lead
RCSB Protein Data Bank
Associate Research Professor
Center for Integrative Proteomics Research
Rutgers, The State University of New Jersey
174 Frelinghuysen Rd
Piscataway, NJ 08854-8087

Email: jasm...@rcsb.rutgers.edu
Phone: (848)445-0103 ext 4920
Fax: (732)445-4320
===

[ccp4bb] job posting: Data Science Engineer

[Michelle Ottaviano]

A joint project between the SBGrid Consortium at Harvard Medical School and
the Dataverse Team at the Institute for Quantitative Social Science at
Harvard University has an immediate opening for a developer to help us
build a next generation data publication system for large biomedical
datasets.

We aim to make biomedical datasets publicly available through a federated
data grid to facilitate access, citation, and data analysis by scientists.
Our pilot collection includes datasets generated using X-ray
crystallography, computer modeling, lattice light sheet microscopy, and
microED diffraction. This collection is currently replicated to computing
centers in the US, Europe, Asia, and South America. The project is
supported by the Helmsley Charitable Trust and was recently selected as a
pilot of the U.S. National Data Service. To learn more about the
environment, please visit our current implementation at data.sbgrid.org and
our group websites at sbgrid.org, slizlab.org, and dataverse.org.

The data science engineer will be embedded within the Dataverse development
team and will primarily be focused on implementing the features necessary
for the successful completion of this project. Examples of features that
must be added to Dataverse include implementation of APIs for
interoperation with components for large (~100 GB) datasets, automatic data
validation pipelines, custom publishing workflows, and other features
relevant to specific biomedical data types. All new functionality developed
under this project will be merged into the Dataverse open source project
and shared with the community.

As a member of our team, this person can expect to collaborate with
researchers, collection specialists, and present outcomes of the project at
meetings and conferences.

Advanced degree (computer science, bioinformatics or engineering preferred)
and 3-5 years of strong programming experience is strongly preferred,
preferably in Java and Python, ideally in the context of web applications.

Our team will welcome candidates with diverse technical backgrounds, but
the successful candidate will have experience handling large datasets and
working as a part of an agile software development team. A working
knowledge of Linux, shell scripting, databases, and distributed version
control systems (git, mercurial, etc) is also necessary. The ideal
candidate will also be familiar with data management software and the
handling and analysis of large datasets.

This is a term appointment ending on September 30, 2018. To apply, e mail
ap...@hkl.hms.harvard.edu.

[ccp4bb] Parallelization?

[Bernhard Rupp]

Hi Developers,

 

I am wondering whether we fully utilize our hardware, and knowing not much
about the finer detail, I'd like to ask a few questions.

 

Most workstations have 6-12 or more cores, and equipped with plenty of cheap
memory and SSD arrays and some overclocking, these machines

are pretty decent. Some programs, like xds-par which I run under Win10 in a
Fedora/RH VM fully use the cores and are blazing fast. 

On WIN10, some like Shelxd also use all the cores.

 

In principle, almost all multi-solution programs should be able to be
parallelized relatively simple, by spawning threads and combining the 

results later (which I could do e.g. with my arp/warp implements).

 

Unfortunately, also some stuff that could be easily run on multiple cores
like phenix multi-conformer refi, epmr, or similar does not,

or not on Windows. Why not and how difficult is that to change?

 

Second, even other programs that have nested loops (and who has not) can be
compiled with e.g. the ifort compiler to use multiple

threads and cores on the i7 series, at least. It is just weird to have say
refmac putter along in one core on a de facto semi-idle workstation. 

 

Is (automated) parallelization via compiler directives feasible, also on
Win, what would it bring, how difficult?

 

Thx, BR

 

--

Bernhard Rupp

Crystallographiae Vindicis Militum Ordo

  http://www.hofkristallamt.org/

  b...@hofkristallamt.org

+1 925 209 7429

+43 767 571 0536

--

Many plausible ideas vanish 

at the presence of thought

--

 

Re: [ccp4bb] Parallelization?

[Kay Diederichs]

Hi Bernhard,

I parallelized ESSENS, SHELXL, CNS and a few other programs, and wrote a paper 
about it (long time ago, in J. Appl. Cryst I think). It's actually not 
difficult to parallelize with OpenMP, but it needs time - developers seemingly 
rather spend their time on other things.

Automatic parallelization by compiler options helps only in the most trivial 
cases. It is my experience that the code must be adapted to reach a useful 
level of parallelization.

best,

Kay

On Thu, 9 Feb 2017 12:33:05 -0800, Bernhard Rupp  
wrote:

>Hi Developers,
>
>
>
>I am wondering whether we fully utilize our hardware, and knowing not much
>about the finer detail, I'd like to ask a few questions.
>
>
>
>Most workstations have 6-12 or more cores, and equipped with plenty of cheap
>memory and SSD arrays and some overclocking, these machines
>
>are pretty decent. Some programs, like xds-par which I run under Win10 in a
>Fedora/RH VM fully use the cores and are blazing fast.
>
>On WIN10, some like Shelxd also use all the cores.
>
>
>
>In principle, almost all multi-solution programs should be able to be
>parallelized relatively simple, by spawning threads and combining the
>
>results later (which I could do e.g. with my arp/warp implements).
>
>
>
>Unfortunately, also some stuff that could be easily run on multiple cores
>like phenix multi-conformer refi, epmr, or similar does not,
>
>or not on Windows. Why not and how difficult is that to change?
>
>
>
>Second, even other programs that have nested loops (and who has not) can be
>compiled with e.g. the ifort compiler to use multiple
>
>threads and cores on the i7 series, at least. It is just weird to have say
>refmac putter along in one core on a de facto semi-idle workstation.
>
>
>
>Is (automated) parallelization via compiler directives feasible, also on
>Win, what would it bring, how difficult?
>
>
>
>Thx, BR
>
>
>
>--
>
>Bernhard Rupp
>
>Crystallographiae Vindicis Militum Ordo
>
>  http://www.hofkristallamt.org/
>
>  b...@hofkristallamt.org
>
>+1 925 209 7429
>
>+43 767 571 0536
>
>--
>
>Many plausible ideas vanish
>
>at the presence of thought
>
>--
>
>
>
>

Re: [ccp4bb] Parallelization?

[George Sheldrick]

Dear Bernhard,

After Kay showed me how to do it by making SHELXL parallel, I was able 
to make SHELXD, ANODE and SHELXT (my current small molecule direct 
methods program) highly parallel. I'm still working on SHELXE because it 
requires major changes. In principle chain tracing is difficult to make 
parallel because you have  to know what you have already traced.


The talk I gave at the 2016 Computing School in Freudenstadt on the 
subject may be downloaded from the SHELX homepage shelx.uni-goettingen.de


Best wishes, George


On 09.02.2017 22:08, Kay Diederichs wrote:

Hi Bernhard,

I parallelized ESSENS, SHELXL, CNS and a few other programs, and wrote a paper 
about it (long time ago, in J. Appl. Cryst I think). It's actually not 
difficult to parallelize with OpenMP, but it needs time - developers seemingly 
rather spend their time on other things.

Automatic parallelization by compiler options helps only in the most trivial 
cases. It is my experience that the code must be adapted to reach a useful 
level of parallelization.

best,

Kay

On Thu, 9 Feb 2017 12:33:05 -0800, Bernhard Rupp  
wrote:


Hi Developers,



I am wondering whether we fully utilize our hardware, and knowing not much
about the finer detail, I'd like to ask a few questions.



Most workstations have 6-12 or more cores, and equipped with plenty of cheap
memory and SSD arrays and some overclocking, these machines

are pretty decent. Some programs, like xds-par which I run under Win10 in a
Fedora/RH VM fully use the cores and are blazing fast.

On WIN10, some like Shelxd also use all the cores.



In principle, almost all multi-solution programs should be able to be
parallelized relatively simple, by spawning threads and combining the

results later (which I could do e.g. with my arp/warp implements).



Unfortunately, also some stuff that could be easily run on multiple cores
like phenix multi-conformer refi, epmr, or similar does not,

or not on Windows. Why not and how difficult is that to change?



Second, even other programs that have nested loops (and who has not) can be
compiled with e.g. the ifort compiler to use multiple

threads and cores on the i7 series, at least. It is just weird to have say
refmac putter along in one core on a de facto semi-idle workstation.



Is (automated) parallelization via compiler directives feasible, also on
Win, what would it bring, how difficult?



Thx, BR



--

Bernhard Rupp

Crystallographiae Vindicis Militum Ordo

 http://www.hofkristallamt.org/

 b...@hofkristallamt.org

+1 925 209 7429

+43 767 571 0536

--

Many plausible ideas vanish

at the presence of thought

--







--
Prof. George M. Sheldrick FRS
Dept. Structural Chemistry,
University of Goettingen,
Tammannstr. 4,
D37077 Goettingen, Germany
Tel. +49-551-39-33021 or -33068
Fax. +49-551-39-22582

Re: [ccp4bb] Parallelization?

[Dyda]

Am I correct that OpenMP parallelized crystallographic software is in Fortran?

While I think there is OpenMP extensions for C and variants, but not so for 
python.
Is this correct?

Fred
***
Fred Dyda, Ph.D.   Phone:301-402-4496
Laboratory of Molecular BiologyFax: 301-496-0201
DHHS/NIH/NIDDK e-mail:fred.d...@nih.gov  
Bldg. 5. Room 303 
Bethesda, MD 20892-0560  URGENT message e-mail: 2022476...@mms.att.net
Google maps coords: 39.000597, -77.102102
http://www2.niddk.nih.gov/NIDDKLabs/IntramuralFaculty/DydaFred
***

Re: [ccp4bb] Parallelization?

[Bernhard Rupp]

I think Intel has it in the parallel studio.

BR

-Original Message-
From: CCP4 bulletin board [mailto:CCP4BB@JISCMAIL.AC.UK] On Behalf Of Dyda
Sent: Thursday, February 9, 2017 1:36 PM
To: CCP4BB@JISCMAIL.AC.UK
Subject: Re: [ccp4bb] Parallelization?

Am I correct that OpenMP parallelized crystallographic software is in
Fortran?

While I think there is OpenMP extensions for C and variants, but not so for
python.
Is this correct?

Fred
***

Fred Dyda, Ph.D.   Phone:301-402-4496
Laboratory of Molecular BiologyFax: 301-496-0201
DHHS/NIH/NIDDK e-mail:fred.d...@nih.gov  
Bldg. 5. Room 303 
Bethesda, MD 20892-0560  URGENT message e-mail: 2022476...@mms.att.net
Google maps coords: 39.000597, -77.102102
http://www2.niddk.nih.gov/NIDDKLabs/IntramuralFaculty/DydaFred

***

Re: [ccp4bb] Composit omit map vs. ligand

[Pavel Afonine]

In addition to excellent Kay's reply..

Also make sure to check refined B factors. Note, if the ligand is not there
then that volume is likely filled with bulk-solvent. Now low occupancy in
combination with very large B factors may approximate bulk-solvent quite
well. The Polder map along with three CC values that its calculation
procedure reports should give you the answer whether the ligand is there or
not.

Pavel

On Wed, Feb 8, 2017 at 2:43 AM, Kay Diederichs <
kay.diederi...@uni-konstanz.de> wrote:

> Dear Petr,
>
> if I understand correcty, the mFo-DFc density (1)  shows almost nothing,
> but the 2mFo-DFc  (2) as well as the composite omit map (3) show the ligand?
>
> As you say, the apparent contradiction between (1) versus (2)&(3) is
> unexpected. One explanation could be that the Fc are simply too bad, i.e.
> the model not good enough to result in useful signal in the difference
> map.  OTOH, that you see the ligand in (2) may be simply model bias, so is
> not meaningful. (3) is hopeful since there is no model bias.
>
> I would suggest to
> - refine the occupancy, to find out why the density is so weak
> - calculate a  (Fo,soak - Fo,native) (4) map with phases from a model
> unbiased by the ligand
> - try a Polder map (5)
>
> - If the occupancy is around 0.5 or higher, that would be a good sign.
> - but if you don't see density in (4) and (5), then your ligand is
> probably not there in any useful amount
>
> I consider (4) as the most sensible method to show presence of the ligand,
> and it should convince reviewers.
>
> HTH,
>
> Kay
>
>
>
> On Wed, 8 Feb 2017 09:05:50 +0100, Petr Kolenko 
> wrote:
>
> >Dear colleagues,
> >
> >we have a dataset with potential enzyme:ligand complex at 2.2 AA
> >resolution. The ligand is very good substrate for the enzyme, we used
> >soaking. We do not see the ligand in the regular difference electron
> >density, only five out of twenty atoms. However, the ligand placed at
> >the active site (model used from structure of a mutant variant) is
> >refined well, giving no negative peaks in difference electron density
> >map and nice observed electron density. I have calculated composit omit
> >map with annealing in Phenix (input model did not contain the ligand)
> >and the electron density for the ligand is there.
> >
> >I have my own opinion, but we are desperate to obtain such data (more
> >than 40 crystals already tested). My question is, would this be proof of
> >presence of the ligand with reduced occupancy? Will this map convince
> >the reviewers? Is there any other way to validate presence of the ligand?
> >
> >Best regards,
> >Petr
>

[ccp4bb] Suggestions for Initial crystal screening

[lujiuwei]

Hi all,


I have purified a protein, the gel filtration and Native-PAGE shows it is very 
homogeneous. There is no problem when I concentrated it to 20mg/mL. However, I 
found above 85% conditions will cause protein precipitation when I try to 
crystallize it  even at 2mg/mL.  I also tried to add 5% glycerol to the protein 
and there is no improvement. In addition, I found when dilute the well solution 
to 50% will not precipitate protein, but 70% or higher will cause. My protein 
is stocked in the buffer 20mM Hepes, pH 7.5, 150mM NaCl,5mM DTT and the PI is 
about 5.5.So, any suggestions are welcome? 


Thanks very much!


Jiuwei

Re: [ccp4bb] Composit omit map vs. ligand

[Nigel Moriarty]

And here is an excellent Polder tutorial

https://www.youtube.com/watch?v=TcTuMJayh5c

you can watch on YouTube in the Phenix Tutorials channel

Cheers

Nigel

---
Nigel W. Moriarty
Building 33R0349, Molecular Biophysics and Integrated Bioimaging
Lawrence Berkeley National Laboratory
Berkeley, CA 94720-8235
Phone : 510-486-5709 Email : nwmoria...@lbl.gov
Fax   : 510-486-5909   Web  : CCI.LBL.gov

On Thu, Feb 9, 2017 at 3:45 PM, Pavel Afonine  wrote:

>
> In addition to excellent Kay's reply..
>
> Also make sure to check refined B factors. Note, if the ligand is not
> there then that volume is likely filled with bulk-solvent. Now low
> occupancy in combination with very large B factors may approximate
> bulk-solvent quite well. The Polder map along with three CC values that its
> calculation procedure reports should give you the answer whether the ligand
> is there or not.
>
> Pavel
>
>
> On Wed, Feb 8, 2017 at 2:43 AM, Kay Diederichs <
> kay.diederi...@uni-konstanz.de> wrote:
>
>> Dear Petr,
>>
>> if I understand correcty, the mFo-DFc density (1)  shows almost nothing,
>> but the 2mFo-DFc  (2) as well as the composite omit map (3) show the ligand?
>>
>> As you say, the apparent contradiction between (1) versus (2)&(3) is
>> unexpected. One explanation could be that the Fc are simply too bad, i.e.
>> the model not good enough to result in useful signal in the difference
>> map.  OTOH, that you see the ligand in (2) may be simply model bias, so is
>> not meaningful. (3) is hopeful since there is no model bias.
>>
>> I would suggest to
>> - refine the occupancy, to find out why the density is so weak
>> - calculate a  (Fo,soak - Fo,native) (4) map with phases from a model
>> unbiased by the ligand
>> - try a Polder map (5)
>>
>> - If the occupancy is around 0.5 or higher, that would be a good sign.
>> - but if you don't see density in (4) and (5), then your ligand is
>> probably not there in any useful amount
>>
>> I consider (4) as the most sensible method to show presence of the
>> ligand, and it should convince reviewers.
>>
>> HTH,
>>
>> Kay
>>
>>
>>
>> On Wed, 8 Feb 2017 09:05:50 +0100, Petr Kolenko <
>> petr.kole...@fjfi.cvut.cz> wrote:
>>
>> >Dear colleagues,
>> >
>> >we have a dataset with potential enzyme:ligand complex at 2.2 AA
>> >resolution. The ligand is very good substrate for the enzyme, we used
>> >soaking. We do not see the ligand in the regular difference electron
>> >density, only five out of twenty atoms. However, the ligand placed at
>> >the active site (model used from structure of a mutant variant) is
>> >refined well, giving no negative peaks in difference electron density
>> >map and nice observed electron density. I have calculated composit omit
>> >map with annealing in Phenix (input model did not contain the ligand)
>> >and the electron density for the ligand is there.
>> >
>> >I have my own opinion, but we are desperate to obtain such data (more
>> >than 40 crystals already tested). My question is, would this be proof of
>> >presence of the ligand with reduced occupancy? Will this map convince
>> >the reviewers? Is there any other way to validate presence of the ligand?
>> >
>> >Best regards,
>> >Petr
>>
>
>

Re: [ccp4bb] Parallelization?

[Robert Oeffner]

Phaser uses OpenMP and is written entirely in C++.

OpenMP isn't available in python. My understanding is that since python itself 
is a language built on top of C an OpenMP loop would have to be implemented at 
the C level anyway. In which case you might as well do it yourself in C, C++ or 
Fortran given that python wasn't designed for CPU intensive tasks.

Rob


Sent from my Windows 10 phone
-- 
Robert Oeffner, Ph.D.
Research Associate, The Read Group
Department of Haematology,
Cambridge Institute for Medical Research
University of Cambridge
Cambridge Biomedical Campus
Wellcome Trust/MRC Building
Hills Road
Cambridge CB2 0XY

www.cimr.cam.ac.uk/investigators/read/index.html
tel: +44(0)1223 763234
mobile: +44(0)7712 887162

From: Dyda
Sent: 09 February 2017 22:09
To: CCP4BB@JISCMAIL.AC.UK
Subject: Re: [ccp4bb] Parallelization?

Am I correct that OpenMP parallelized crystallographic software is in Fortran?

While I think there is OpenMP extensions for C and variants, but not so for 
python.
Is this correct?

Fred
***
Fred Dyda, Ph.D.   Phone:301-402-4496
Laboratory of Molecular BiologyFax: 301-496-0201
DHHS/NIH/NIDDK e-mail:fred.d...@nih.gov  
Bldg. 5. Room 303 
Bethesda, MD 20892-0560  URGENT message e-mail: 2022476...@mms.att.net
Google maps coords: 39.000597, -77.102102
http://www2.niddk.nih.gov/NIDDKLabs/IntramuralFaculty/DydaFred
***