Re: [ccp4bb] offtopic__which crystals to harvest

[Van Den Berg, Bert]

True. It also appears that most people will be looking for less expensive ways 
to distinguish between protein and salt crystals
 
Bert van den Berg
University of Massachusetts Medical School
Program in Molecular Medicine



From: CCP4 bulletin board on behalf of V. Nagarajan
Sent: Wed 1/14/2009 2:49 PM
To: CCP4BB@JISCMAIL.AC.UK
Subject: Re: [ccp4bb] offtopic__which crystals to harvest


Actually, UV fluorescence imaging appears to be a pretty reliable means of
discriminating protein crystals, provided your protein has Trp residue(s).

V. Nagarajan
JAN Scientific, Inc.
http://www.janscientific.com  

-Original Message-
From: CCP4 bulletin board [mailto:ccp...@jiscmail.ac.uk] On Behalf Of Artem
Evdokimov
Sent: Tuesday, January 13, 2009 8:10 PM
To: CCP4BB@JISCMAIL.AC.UK
Subject: Re: [ccp4bb] offtopic__which crystals to harvest

Hi,



A few simple hints:



(Please note that I am aware of the inexact language in the statements below
but I don't have the time to write this up exactly - conversational English
would have to do. Caveat emptor.)



Most protein crystals will break or deform when poked with a steel needle.
Most inorganic salts/minerals won't deform from a simple poke, but will
break (often with an audible crack) when pushed hard against something solid
(like the surface of the glass slip etc.). Many, but not all protein
crystals can survive gentle prodding with a thin cat whisker. Nylon loops
are a bit trickier because they can exhert different forces depending on
their geometry, age of the loop, and user's manual aptitude.



To make matters more complicated - crystals of organic materials (i.e. not
salt but also not protein) can display properties similar to either (but
will more often than not tend to behave like salts).



Salt crystals sink very rapidly in most well solutions. Protein crystals
often take their time (lower density). Salt crystals often display Newton
rings (Newton rainbows) when viewed through a polarizer-analyzer pair (not
to be confused with relatively simple gradients of birefringence colors that
are also common to protein crystals!). I have seen a few crystals of
proteins that had distinct Newton rings and they were all exceptionally good
diffractors. Don't be confused by rainbow-like coloring that's often
associated with spherolites - the latter aren't likely to diffract X-rays in
a useful manner :-)



If in doubt - stick your crystals into an X-ray beam. Pretty much the best
way to resolve this ambiguity! The next best choice is to show the crystals
to an experienced crystallographer - oftentimes it's possible to guess just
by eyeballing the drops but it takes experience to learn the traits and
habits. Membrane crystals (or for that matter any crystals grown in the
presence of detergent) can be extremely tricky to identify correctly due to
the inherently soft and nasty nature of detergent crystals and the tendency
of the latter to form various quasi-crystalline artefacts.



Good luck,



Artem



From: CCP4 bulletin board [mailto:ccp...@jiscmail.ac.uk] On Behalf Of
deliang
Sent: Tuesday, January 13, 2009 10:50 PM
To: CCP4BB@JISCMAIL.AC.UK
Subject: [ccp4bb] offtopic__which crystals to harvest



Hi there,



Since a lot of different forms of crystals shows, I am using a quick/simple
strategy to choose crystals by applying a force on the crystal against the
wall, with the nylon loop. 



Some can never break apart, so they are salt crystals? The others can not
survive the force and lose their intact shape and sharp surface. It seems
these are protein crystals, but are they "bad" crystals?  I just came to
this field, and welcome all your suggestions and experience.



Thanks a lot.



Deliang

Re: [ccp4bb] offtopic__which crystals to harvest

[V. Nagarajan]

Actually, UV fluorescence imaging appears to be a pretty reliable means of
discriminating protein crystals, provided your protein has Trp residue(s).

V. Nagarajan
JAN Scientific, Inc.
http://www.janscientific.com

-Original Message-
From: CCP4 bulletin board [mailto:ccp...@jiscmail.ac.uk] On Behalf Of Artem
Evdokimov
Sent: Tuesday, January 13, 2009 8:10 PM
To: CCP4BB@JISCMAIL.AC.UK
Subject: Re: [ccp4bb] offtopic__which crystals to harvest

Hi,

 

A few simple hints:

 

(Please note that I am aware of the inexact language in the statements below
but I don't have the time to write this up exactly - conversational English
would have to do. Caveat emptor.)

 

Most protein crystals will break or deform when poked with a steel needle.
Most inorganic salts/minerals won't deform from a simple poke, but will
break (often with an audible crack) when pushed hard against something solid
(like the surface of the glass slip etc.). Many, but not all protein
crystals can survive gentle prodding with a thin cat whisker. Nylon loops
are a bit trickier because they can exhert different forces depending on
their geometry, age of the loop, and user's manual aptitude.

 

To make matters more complicated - crystals of organic materials (i.e. not
salt but also not protein) can display properties similar to either (but
will more often than not tend to behave like salts).

 

Salt crystals sink very rapidly in most well solutions. Protein crystals
often take their time (lower density). Salt crystals often display Newton
rings (Newton rainbows) when viewed through a polarizer-analyzer pair (not
to be confused with relatively simple gradients of birefringence colors that
are also common to protein crystals!). I have seen a few crystals of
proteins that had distinct Newton rings and they were all exceptionally good
diffractors. Don't be confused by rainbow-like coloring that's often
associated with spherolites - the latter aren't likely to diffract X-rays in
a useful manner :-)

 

If in doubt - stick your crystals into an X-ray beam. Pretty much the best
way to resolve this ambiguity! The next best choice is to show the crystals
to an experienced crystallographer - oftentimes it's possible to guess just
by eyeballing the drops but it takes experience to learn the traits and
habits. Membrane crystals (or for that matter any crystals grown in the
presence of detergent) can be extremely tricky to identify correctly due to
the inherently soft and nasty nature of detergent crystals and the tendency
of the latter to form various quasi-crystalline artefacts.

 

Good luck,

 

Artem



From: CCP4 bulletin board [mailto:ccp...@jiscmail.ac.uk] On Behalf Of
deliang
Sent: Tuesday, January 13, 2009 10:50 PM
To: CCP4BB@JISCMAIL.AC.UK
Subject: [ccp4bb] offtopic__which crystals to harvest

 

Hi there,

 

Since a lot of different forms of crystals shows, I am using a quick/simple
strategy to choose crystals by applying a force on the crystal against the
wall, with the nylon loop.  

 

Some can never break apart, so they are salt crystals? The others can not
survive the force and lose their intact shape and sharp surface. It seems
these are protein crystals, but are they "bad" crystals?  I just came to
this field, and welcome all your suggestions and experience.

 

Thanks a lot.

 

Deliang

Re: [ccp4bb] Ramachandran plot, in a text list

[Edward A. Berry]

John Pak wrote:
Hi, sorry if this was posted earlier.  How can I calculate a 
Ramachandran plot, but output the information to a text file list?  
i.e.  something like Ala51, phi=X, psi=Y.


I can't seem to find this information in the ccp4 SFCheck/Procheck log 
file.




And lets not forget procheck, since its included with ccp4 and you
probably already have it. After you run procheck on your protein
(procheck pdbname.pdb 1.8), there is a text file pdbname.rin
which lists phi, psi, omega and sidechain dihedrals for each residue:


  phipsi   omega  chi1   chi2   chi3chi4
  14ASN A  15 E -89.06 141.45-179.73 -63.59 -55.39 999.90 999.90  -2.14   0.00  
34.00  89.18 88.838 88.830  8 45  1.980  0.762
  15VAL A  16 E-139.67 138.33 177.89-170.03 999.90 999.90 999.90   0.00   0.00  
36.23  87.91 86.643 87.253  8 50  0.147  0.499
  16THR A  17 E-144.59 139.86-179.12  51.94 999.90 999.90 999.90  -2.64   0.00  
34.56  84.27 86.238 85.020  8 40  0.421  0.536

You also have a .ps file pdbname_01.ps with the graphical ramachandran
which you can view with gv or gs, or convert to pdf with ps2pdf

DSSP also list phi-psi angles, in two columns far to the right.

Ed

Re: [ccp4bb] Ab initio structure prediction

[Jürgen Bosch]

Hi Riya,

you could use Modeller , Robetta http://robetta.bakerlab.org or check  
out the Expasy server what they offer in terms of threading programs http://www.expasy.ch/tools/#tertiary


Jürgen
On 14 Jan 2009, at 13:08, riya doreen wrote:


Dear All,

I will be very grateful if someone can point me to a program for ab  
initio structure prediction of small peptide fragments (50-100 amino  
acids).


Thanks
Riya


-
Jürgen Bosch
Johns Hopkins Bloomberg School of Public Health
Biochemistry and Molecular Biology, W8708
615 North Wolfe Street
Baltimore, MD 21205
Phone: +1-410-614-4742

Re: [ccp4bb] Ramachandran plot, in a text list

[Robert Immormino]

Hi John,

Another way to get at this data is to run your structure through
MolProbity (http://molprobity.biochem.duke.edu/).  After this on the
mainpage there in the downloads section...if you go to *all
downloads*... there is a raw_data section that contains the
ramachandran stats (PDBid-rama.data ) and other validation statistics
as colon delimited data files.

Cheers,
-bob

On Wed, Jan 14, 2009 at 10:05 AM, John Pak  wrote:
> Hi, sorry if this was posted earlier.  How can I calculate a Ramachandran
> plot, but output the information to a text file list?  i.e.  something like
> Ala51, phi=X, psi=Y.
>
> I can't seem to find this information in the ccp4 SFCheck/Procheck log file.
>

Re: [ccp4bb] summary: crystals from skin

[Jesse]

Are the crystals growing as a result of the skin, or coincidentally located
with the skin?  For instance, if you are using the hanging drop method, and
the crystals are growing with the skin because of their density, you could
switch to sitting drop plates -- crystals grow on the bottom, skin stays on
top.  If the skin is required for crystal growth, then of course this
doesn't apply.
Generally, this is also simpler than converting the growth conditions from
VD to batch under oil.

-Jesse

On Tue, Dec 30, 2008 at 11:22 AM, riya doreen  wrote:

> Hello Everyone,
>
> Thank you for all the suggestions on how to manipulate skin on crystal
> drops. Here is a summary of responses:
>
> (i) Higher concentrations of reducing agents (ii) Layering crystal drop
> with cryo-protectant (iii) Trypsinize the skin (iv) Grow crystals under oil
> (v) Try 4C if crystals are grown at 18 or RT (vi) Lower  protein
> concentration and (vii) Fiddling around with the drop which may break the
> skin and free the crystals
>
>
> On Sun, Dec 28, 2008 at 8:24 PM, riya doreen  wrote:
>
>> *Hello All,
>>
>> I am trying to optimize protein crystals that grow from skin. All efforts
>> to separate crystals from skin have failed since it is fairly thick.
>> Crystallization under oil eliminates skin but the resultant crystals are
>> very thin and small needles.
>>
>> I would appreciate any suggestions on how this problem can be eliminated.
>>
>> Thanks
>>
>> *
>
>
>

Re: [ccp4bb] Ramachandran plot, in a text list

[Pavel Afonine]

Hi,

you can do it PHENIX (http://www.phenix-online.org/):

phenix.ramalize model.pdp

and it will give you this list.

Pavel.

PS> You need to have the latest version of PHENIX for this.


On 1/14/2009 10:05 AM, John Pak wrote:
Hi, sorry if this was posted earlier.  How can I calculate a 
Ramachandran plot, but output the information to a text file list?  
i.e.  something like Ala51, phi=X, psi=Y.


I can't seem to find this information in the ccp4 SFCheck/Procheck log 
file.

[ccp4bb] Ab initio structure prediction

[riya doreen]

Dear All,

I will be very grateful if someone can point me to a program for ab initio
structure prediction of small peptide fragments (50-100 amino acids).

Thanks
Riya

[ccp4bb] Ramachandran plot, in a text list

[John Pak]

Hi, sorry if this was posted earlier.  How can I calculate a  
Ramachandran plot, but output the information to a text file list?   
i.e.  something like Ala51, phi=X, psi=Y.


I can't seem to find this information in the ccp4 SFCheck/Procheck log file.

[ccp4bb] Structural Biology at the Society for General Microbiology Meeting, 30th March- 2nd April 2009

[Arwen Pearson]

(Posted on behalf of Nic Stonehouse)

The forthcoming Society for General Microbiology Meeting includes a 
symposium on ‘Structural Insights into Virus Biology’, with 
presentations from leading structural biologists. Also included is a 
work shop session entitled ‘Virus Structure’, where PhD students and 
post-doctoral fellows are encouraged to present their work . The meeting 
runs from 30 March - 2nd April 2009 in Harrogate, UK, with the 
structural symposium on 1 and 2 April.


Information on the meeting can be found at 
http://www.sgm.ac.uk/meetings/MTGPAGES/Har0919.cfm
with details on abstract submission at 
http://www.sgm.ac.uk/meetings/MTGPAGES/Har09ViWshopsSpec.cfm.


The deadline for abstracts is 23 January (submission to 
n.j.stoneho...@leeds.ac.uk) but conference registration is open until 
March. There are generous bursaries available for PhD students who are 
members of the Society.

--

Dr Arwen R Pearson
RCUK Academic Research Fellow
Astbury Centre for Structural Molecular Biology
Astbury Building
The University of Leeds
LS2 9JT
United Kingdom

(t) +44 (0)113 34 33032
(e) a.r.pear...@leeds.ac.uk

[ccp4bb] unknown PyMOL script author

[MARTYN SYMMONS]

Dear CCP4ers
Do you recognize your Python code in the script below? I got this from a script 
in the PyMOL wiki last year but now I can't locate it or identify the author. I 
have used this syntax extensively in some structural analysis that I am now 
writing up and I would like to credit the originator (PyMOL and the wiki both 
get a citation obviouslybut I checked and neither Warren Delano nor Jason 
Vertrees is the author). 
 
for a in cmd.index("A//37/CB"):
for b in cmd.index("B//LYS/NZ"):
cmd.select("s1","%s`%d"%a)
cmd.select("s2","%s`%d"%b)
if cmd.select("(s1|s2) and not ?skip"):
if cmd.dist("tmp","s1","s2") < 20.0 :
  print '',cmd.iterate("s1|s2","print 
'',segi,resn,resi,name"),round(cmd.dist("tmp","s1","s2"),3)

Thanks
Best wishes
  Martyn 

Martyn Symmons
Department of Pathology
University of Cambridge

Re: [ccp4bb] Refmac library trouble -

[junfeng liu]

Hi Jon,
I do not think the lib you got from ligand dep is enough to be used in 
the refmac.
Unfortunately  I can not  find or make  the  library from PRODRG2 server 
because the FE can not be handled by that server right now.
So maybe you should make the  library yourself : either using  sketcher 
- monomer library sketcher  in CCP4 or elbow in PHENIX .

Best wishes!
leo

[ccp4bb] polls on ccp4bb

[Martyn Winn]

This particular poll is fairly innocuous, but it does remind me to make
a general comment on polls on ccp4bb. There seems to be an increasing
trend of running polls on ccp4bb, some of them getting very specific
with regard to program or instrument or manufacturer. 

While it is certainly useful to gather other people's experiences, there
is a tendency to draw conclusions not backed by evidence (low number of
responses, not comparing like with like, etc.). It has been suggested
that scientists are not supposed to do this 

In fact, we have received objections to some polls that have been run
here. I think the rule is, it is ok to gather people's experiences, but
if you want to draw firm conclusions you need to do a proper
peer-reviewed study.

As we wish ccp4bb to remain self-policing, there is an etiquette section
at http://www.ccp4.ac.uk/ccp4bb.php

Cheers
Martyn

On Wed, 2009-01-14 at 16:13 +0900, Ngo Duc Tri wrote:
> Dear Crystallography Scientists,
> 
> We are working on some membrane proteins and would like to apply
> monoclonal antibody to facilitate the crystallization. Even this
> method showed very good results in some cases, I am really not sure
> about other failure cases that are not reported.
> 
> Because this method is quite expensive and labourous, having the
> statistics of the failures is quite helpful for researchers to decide
> what they should do. That's the reason I would like to ask you to give
> some minutes to make a poll. Your information is really helpful for
> other researchers who are considering to apply this method to their
> project.
> 
> Thank you so much for your time!
> 
> Voter: Any researchers who are working on membrane protein.
> Please follow this link to submit your vote.
> http://smsb1.skku.ac.kr/pages/mypoll/
> 
> With best regards,
> TriNgo,
> PhD Student,
> School of Medicine - Sungkyukwan University, Korea
> Lab: http://smsb1.skku.ac.kr/
-- 
***
* *
*   Dr. Martyn Winn   *
* *
*   STFC Daresbury Laboratory, Daresbury, Warrington, WA4 4AD, U.K.   *
*   Tel: +44 1925 603455E-mail: martyn.w...@stfc.ac.uk*
*   Fax: +44 1925 603825Skype name: martyn.winn   * 
* URL: http://www.ccp4.ac.uk/martyn/  *
***

[ccp4bb] CALL FOR PROPOSALS FOR ESRF BEAMTIME WITH ONLINE MICROSPEC

[nurizzo]

CALL FOR PROPOSALS FOR BEAMTIME WITH ONLINE MICROSPEC

Proposal Deadline 31st January 2009

There will be beamtime available at the ESRF for MX data collection with
a setup that allows online monitoring of UV/VIS spectral changes of the
crystal during the X-ray diffraction experiment. Users who are
interested in using this beam time (including those who are members of
BAG Groups) should use the following mechanism:

http://www.esrf.fr/UsersAndScience/UserGuide/Applying/ProposalGuidelines/MXnon-BAGproposal 



and it must be clearly indicated in the title of the proposal form that
the online monitoring of spectral changes is necessary for the project.

A brief description of the device is given below however users are
encouraged to consult the web pages for detailed information:

http://www.esrf.fr/UsersAndScience/Experiments/MX/How_to_use_our_beamlines/Run_Your_Experiment/Microspectrophotometer_User_Guide 



As this is not a standard setup, it might take a significant amount of
time to train users, align the device, and analyse the data in order to
derive relevant data collection schemes. We will therefore schedule 24
hours for each project. The deadline for this specific application is
Saturday 31st January 2009


*Dates of beam-time: 20th - 23rd February 2009
*Storage Ring: 7/8 +1 (200mA)
Beamline: ID14-2
Energy: 13.29 keV

Specifications:
UV/VIS-range: 250-800nm
ODmax: 2-2.5
min integration time: 50ms
Light source: OceanOptics DH2000, Halogen/D2-lamp
Monitoring Light size: 0.03 (min) - 0.15mm(max)
Sampling freq (to disk): 10Hz or lower



--
Dr Didier NURIZZO
MX Beam-Line Operations Manager  Tel : (+33) 4 76 88 29 00
Macromolecular Crystallography Group Fax : (+33) 4 76 88 26 24
ESRF
B.P. 220, 6 rue Jules Horowitz   e-mail : nuri...@esrf.fr
F-38043 GRENOBLE CEDEX   http://www.esrf.fr