Re: [ccp4bb] problem in transformation of pqe 30 clone

2009-07-17 Thread Dianfan Li

Hi,

I only read the topic this morning -

In some cases, I have to wait 30+ hours to see colonies when I transform 
BL21 DE3 Star while normally they should appear after 14 hours.


Dianfan

ar...@xtals.org wrote:

Hello

In order to know *exactly* what the reason for poor transformation outcome
was one has to do all sorts of experiments. This is not likely to be your
goal, right? Leaky expression, overload of DNA, somehow compromised cells,
or even plain old user error - and numerous other reasons can be proposed
(including some esoteric ones like incompatibility of your gene product
with a specific strain, for reasons going beyond protocol errors and
simple considerations).

This can take a couple of years to sort out.

If you're talking about using M15[pREP4] cells then the reason why you
failed earlier is likely to be toxicity - pREP4 plasmid supplies lacI
in-trans, which represses 'stray' transcription reasonably well. The cells
that you used before and did not succeed with must have had low intrinsic
levels of lacI and thus permitted leakage.

To test this theory you could streak a couple of colonies of the
M15[pREP4] cells on a plate that has a small amount of IPTG in the agar
(say 0.1 mM) - if the colonies don't grow at all, or appear to be very
tiny then you know it's your protein toxicity that's the issue here. Keep
in mind that bacteria are extremely survival-oriented and therefore you
will eventually generate 'safe' mutants and recombinants that will be able
to grow even on IPTG. Likely those mutants will be useless to you from
practical standpoint.

Once you induce the M15[pREP4] cells - the toxicity will manifest itself
again. You will likely observe total cessation of growth (for a long
period of time) and possibly even lysis of the culture. Therefore you may
have to adjust your fermentation protocol to take this into account. One
of the common adjustments is deliberate induction at higher OD, another is
to include 0.8% glucose in the growth medium; there are other options
available to you. Hopefully once you induce transcription, the gene
products do not shut down protein synthesis - which would be a disaster
since it would likely shut down its own synthesis as well. Since I don't
know what you're growing, this is just one of the possibilities :)

Artem

  

now i have transformed pqe 30 clone into the m15 host successfully, can
someone let me know exactly what was reason behind the problem into
transformation??is it leaky expression into the other host that is toxic
for the cells,if it so then will i would be  able to get good expression
into this host???

atul
-Original Message-
From: CCP4 bulletin board on behalf of ar...@xtals.org
Sent: Tue 5/5/2009 6:24 AM
To: CCP4BB@JISCMAIL.AC.UK
Subject: Re: [ccp4bb] problem in transformation of pqe 30 clone

To clarify: I am not implying that I've worked with many proteins that
express better in XL1-blue cells than they would express in BL21(DE3) or
such if cloned into a pET vector or similar. In fact I can probably recall
only one or two that expressed *better* in XL1 cells. However in the good
old days pQE series of vectors was quite commonly used and I had things in
that were inherited from others - these 'things' were fairly simple to
express in XL1-blue whereas they gave me loads of trouble in other strains
and I was too busy/lazy to re-clone them.

Artem



Hello Artem,

 We express almost all our proteins in BL21 derivatives. It sounds
like you've worked with many proteins that express/behave better in
XL1-Blue?


ho
UC Berkeley

-
XL1-Blue is a strain of E. coli. Whether it is or isn't an expression
host
depends on the definition, and I am not going to argue semantics.

The T5 promoter works with regular garden variety RNApol of E. coli.
Therefore ANY E. coli strain is an 'expression host' for vectors that
contain this promoter.

I've expressed many proteins in XL1-Blue and I see no reason why you
can't
express yours, either.

Artem

  



Re: [ccp4bb] heavy atom derivative choice

2009-07-17 Thread Kay Diederichs

David Briggs schrieb:



Bart Hazes used to have an excellent set of notes online about heavy
atom derivatisation - I can't seem to find the URL right now...



I believe the CCP4 wiki has the information that you're looking for (at 
http://strucbio.biologie.uni-konstanz.de/ccp4wiki/index.php/Soaking_with_heavy_atoms 
) Now if people would add their notes to the wiki that would actually 
become an even better ressource ...


Unfortunately it seems that Bart Hazes' webpage cannot currently be 
accessed, and the cache results do no longer seem to appear in my 
Google searches!


best,

Kay
--
Kay Diederichshttp://strucbio.biologie.uni-konstanz.de
email: kay.diederi...@uni-konstanz.deTel +49 7531 88 4049 Fax 3183
Fachbereich Biologie, Universität Konstanz, Box M647, D-78457 Konstanz

This e-mail is digitally signed. If your e-mail client does not have the
necessary capabilities, just ignore the attached signature smime.p7s.



smime.p7s
Description: S/MIME Cryptographic Signature


[ccp4bb] NCCR Symposium on New Trends in Structural Biology / NCCR Practical Course on Biomolecular Modelling

2009-07-17 Thread Patrick Sticher

Dear colleagues,

we have two announcements to make:
_

7th International NCCR Symposium on New Trends in Structural Biology
7 + 8 September 2009, ETH Zürich, Lecture Hall HG E7, Zürich, Switzerland
www.structuralbiology.uzh.ch/symposium2009

The lecture program is available online at:
http://www.structuralbiology.uzh.ch/symposium2009/program.asp

Plenary lecturers:
Frédéric Allain, Pietro de Camilli, Raimund Dutzler, Arthur L. Horwich, 
Brian Kobilka, Harry Noller,

Anna Marie Pyle, Ben Schuler, David Wemmer, Masasuke Yoshida

Online registration to this event is still possible through the 
symposium homepage or directly at:

http://www.structuralbiology.uzh.ch/symposium2009/registration09.asp
_

8th NCCR Practical Course and 2nd Winter School
Biomolecular Modelling

The 8th NCCR Practical Course will cover key topics in the area of 
computational structural biology. The course format includes morning 
lectures by experts in the field that alternate with later afternoon to 
evening tutorials and discussions. The course will be organized as a 
winter retreat in the Swiss alps offering participants the opportunity 
to learn in a stimulating atmosphere, enjoy winter activities in the 
swiss alps during course breaks, and network and exchange ideas on an 
informal basis with fellow participants, tutors, and lecturers.


More information at http://www.structuralbiology.uzh.ch/course2010.asp
Online Application will be possible as of July 31, 2009
_

Please do not hesitate to contact me anytime if you need further 
information (stic...@bioc.uzh.ch).


With best regards,
Patrick Sticher

The NCCR Structural Biology is a research initiative of the Swiss 
Science Foundation. Its research encompasses the fields of recombinant 
protein technologies, macromolecular structure determination and 
computational biomolecular sciences with a special focus on membrane 
proteins and supramolecular assemblies/interactions. 13 research groups 
from Swiss Universities and Research Institutions participate in this 
network. www.structuralbiology.uzh.ch/


--
_
Visit the NCCR on the Internet
www.structuralbiology.uzh.ch

Dr. Patrick Sticher Moser
NCCR Scientific Officer
Institute of Biochemistry
University of Zürich
Winterthurerstrasse 190
CH - 8057 Zürich

Phone   +41 / (0)44 / 635 54 84
Fax +41 / (0)44 / 635 59 08
Mailstic...@bioc.uzh.ch


[ccp4bb] Postdoc Position at EMBL Grenoble

2009-07-17 Thread Sandor Brockhauser

The Schaffitzel laboratory at the European Molecular Biology Laboratory (EMBL) 
Grenoble, France seeks to recruit an outstanding postdoctoral scientist in 
structural biology with a research focus directed towards the structure of 
macromolecular assemblies. The major theme within the group is structure and 
function of ribosomal complexes, co-translational targeting and translocation 
(Schaffitzel et al., Nature 2006; Mitra, Schaffitzel et al., Nature 2005; 
Kohler et al., Mol.Cell 2009). More information:
http://www.embl.fr/research/unit/schaffitzel/index.html

This position requires a Ph.D. in biochemistry or a related field, with a strong background in protein biochemistry. Applicant should have experience in guiding structure determination projects from start to finish, experience in expressing and purifying membrane proteins and/or electron microscopy would be advantageous. The candidate should be a highly motivated individual who enjoys working as part of a collaborative and multidisciplinary team. 


The laboratory is well-situated in a structural biology environment at the 
Polygone Scientifique in Grenoble. End of 2009, we will be equipped with a 
state-of-the-art electron microscope (Polara, FEI). Access to modern 
biophysical instrumentation (analytical ultracentrifugation, surface Plasmon 
resonance, dynamic light scattering, isothermal calorimetry, CD spectrometer) 
is provided.

The position is available from January 2010 and funded for 2 years.
It may be extended dependent on performance.

To apply please send your CV, a statement of research interests, and names (including 
email address) of at least two referees by email to Dr. Christiane Schaffitzel 
(schaffit...@embl.fr mailto:schaffit...@embl.fr).
Applications will be accepted until the position is filled.


[ccp4bb] Keystroke changes on Coot

2009-07-17 Thread sharotka godzina







I obtained Coot using Fink, and I'm on a MacBook Pro running OS 10.5.7.  I've 
found that in Coot, but in no other programs, my keystrokes are altered.  Space 
bar unzooms, 2 reads r, and delete returns a ,.  Downloading from 
William Scott's webpage didn't help.  Any suggestions?

-Sarah

_
Windows Live™ SkyDrive™: Get 25 GB of free online storage.
http://windowslive.com/online/skydrive?ocid=TXT_TAGLM_WL_SD_25GB_062009

Re: [ccp4bb] Keystroke changes on Coot

2009-07-17 Thread Murray, Michael (NIH/NIEHS)
I had this problem earlier in the week. Apparently it's recently become an
issue.

I found the answer here
http://www.techsupportforum.com/alternative-computing/mac-support/389812-x1
1-keyboard-map-messed-up.html

Briefly, there seems to be a problem with they keyboard mapping in X11.  You
probably see the same thing in other X11 programs like CCP4.  If you delete
/usr/X11/share/X11/xkb the problem should go away.

Good luck,

Mike

Michael S. Murray, Ph.D.
National Institute of Environmental Health Sciences
MD E3-01
P.O. Box 12233
Research Triangle Park, NC 27709
Phone: (919) 541-0268


On 7/17/09 11:06 AM, sharotka godzina sgodz...@hotmail.com wrote:

 I obtained Coot using Fink, and I'm on a MacBook Pro running OS 10.5.7.  I've
 found that in Coot, but in no other programs, my keystrokes are altered.
 Space bar unzooms, 2 reads r, and delete returns a ,.  Downloading
 from William Scott's webpage didn't help.  Any suggestions?
 
 -Sarah
 
 
 Windows Live SkyDrive: Get 25 GB of free online storage.   Get it on your
 BlackBerry or iPhone.
 http://windowslive.com/online/skydrive?ocid=TXT_TAGLM_WL_SD_25GB_062009 


Re: [ccp4bb] OpenGL Stereo 3D on 120 Hz LCDs, at last!

2009-07-17 Thread Jim Fairman
Warren,
Do you have a link to the beta drivers?


On Fri, Jul 17, 2009 at 1:20 AM, Warren DeLano war...@delsci.com wrote:

  FYI, for folks not subscribed to pymol-users:



 nVidia today released beta drivers which at last enable OpenGL-based stereo
 3D visualization on 120 Hz LCDs using Quadro graphics cards.  So long as you
 are willing to put up with Windows, you can finally abandon those old CRTs
 without spending a fortune and without sacrificing quality of the stereo 3D
 effect.



 Details posted at http://www.pymol.org



 Cheers,

 Warren






-- 
Jim Fairman
Graduate Research Assistant
Department of Biochemistry, Cellular, and Molecular Biology (BCMB)
University of Tennessee -- Knoxville
216-368-3337 jfair...@utk.edu james.fair...@case.edu


[ccp4bb] Scalepack error model?

2009-07-17 Thread Richard Gillilan
Does anyone know of a detailed rigorous discussion of how the  
scalepack error model/Bayesian reasoning works? The scalepack manual  
has no equations for this.


Richard Gillilan
MacCHESS


[ccp4bb] What is a Macromolecular Complex? Amsterdam, 1-2 October 2009

2009-07-17 Thread Anastassis Perrakis

Dear all,

The EU FP7 programs SPINE-2-Complexes,  3d-Repertoire and TEACH-SH are  
organizing a joined workshop about:


What is a Macromolecular Complex?

As the subtitle implies, we want to look at the Shades of Meaning  
across Cellular, Systems and Structural Biology.


We have what we believe that is an excellent collection of speakers,  
Philippe Bastiaens (Dortmund), Gianni Cesareni (Rome)
Elena Conti (Martinsried), Stephen Cusack (Grenoble), Jan Ellenberg  
(Heidelberg),Toby Gibson (Heidelberg),
Christina Kiel (Barcelona),Andreas Ladurner (Heidelberg), Ohad Medalia  
(Beer-Sheva), Dino Moras (Illkirch)
Andrea Musacchio (Milano), Laurence Pearl (London), Bertrand Seraphin  
(Gif-sur-Yvette)
Holger Stark (Gottingen), Ulrich Stelzl (Berlin), Dave Stuart  
(Oxford), Peter Tompa (Budapest),
The exact talks will travel across many scientific subjects covering  
cell, molecular, structural, computation and systems biology,
and we hope to put together a lively and useful two-day workshop, that  
will help to shape interests, interactions, and approaches for the  
future.


The workshop will take two days, 1-2 October, will take place in  
Amsterdam with all its cultural amenities,

and is free for everybody to attend, following registration.

Any such workshop would be nothing without keen and lively  
participants, so we invite you to have a look at


http://xtal.nki.nl/Oct2009

for details on the program, dates, etc and - we hope - to register soon!

Last but not least, our Dutch readers, should check the note about the  
Protein Facility opening event in the morning of the 1st of October.


best regards,

Stephen Cusack
Andrea Musacchio
Carlo Petosa
Anastassis Perrakis


[ccp4bb] EPMR statistics

2009-07-17 Thread Jerry McCully

Dear ALL:

  We are trying to do a molecular replacement using EPMR. IT is weird that 
we got high CC(0.56) but with high R factor about 0.89 as well.

   How can we optimize our EPMR trials to get a definitive correct 
solutions?

   Thanks a lot for the help.

Have a nice weekend!

Jerry McCully

 

_
Lauren found her dream laptop. Find the PC that’s right for you.
http://www.microsoft.com/windows/choosepc/?ocid=ftp_val_wl_290

Re: [ccp4bb] Scalepack error model?

2009-07-17 Thread Dominika Borek

Described in:

Borek, D., Minor, W.  Otwinowski, Z. (2003) Measurement errors and 
their consequences in protein crystallography. Acta Cryst. D59: 2031-2038.


http://scripts.iucr.org/cgi-bin/paper?ba5035

(open access) or

http://bones.swmed.edu/pdf/7_Borek_et_al_ActaCrystD_2003_D59_2031-2038%5B1%5D.pdf

Dominika

Does anyone know of a detailed rigorous discussion of how the  
scalepack error model/Bayesian reasoning works? The scalepack manual  
has no equations for this.






--


Dominika Borek
UT Southwestern Medical Center
5323 Harry Hines Boulevard
Room ND10.214
Dallas, Texas 75390-8816
phone: +1 214-645-6378
fax: +1 214-645-6353
domin...@work.swmed.edu


Re: [ccp4bb] Scalepack error model?

2009-07-17 Thread Joe Cockburn
Dear Richard,
I *think* it works like this, don't know if it's detailed or rigourous
enough for you!

If I(l,h,i) is the intensity of the ith observation of reflection h on
frame l, with error sig(l,h,i), and S(l) is the (inverse) scale factor to
be applied to frame l, then the error in the scaled intensity
I(l,h,l)/S(l) is parameterised in terms of the error scale factor (E1) and
estimated error (E2) as

sqrt( (E1*sig(l,h,i))**2  +  (E2*I(h)*S(l))**2  )

where I(h) is the weighted mean of the scaled intensity values for index h
(i.e. the merged, scaled intensity).

The reasoning behind this that the errors in the intensities of strong and
weak reflections generally arise from different sources. Weak data are
noisy, whilst very strong data can often be systematically badly measured,
especially in DENZO, which assumes that all your spots are the same size
and shape, or overloaded. E1 and E2 thus tend to dominate the error model
at high and low resolutions, respectively.
Hope that helps,
Joe


 Does anyone know of a detailed rigorous discussion of how the
 scalepack error model/Bayesian reasoning works? The scalepack manual
 has no equations for this.

 Richard Gillilan
 MacCHESS



Re: [ccp4bb] Scalepack error model?

2009-07-17 Thread Joe Cockburn
Dear Richard,
I *think* it works like this, don't know if it's detailed or rigourous
enough for you!

If I(l,h,i) is the intensity of the ith observation of reflection h on
frame l, with error sig(l,h,i), and S(l) is the (inverse) scale factor to
be applied to frame l, then the error in the scaled intensity
I(l,h,l)/S(l) is parameterised in terms of the error scale factor (E1) and
estimated error (E2) as

sqrt( (E1*sig(l,h,i))**2  +  (E2*I(h)*S(l))**2  )

where I(h) is the weighted mean of the scaled intensity values for index h
(i.e. the merged, scaled intensity).

The reasoning behind this that the errors in the intensities of strong and
weak reflections generally arise from different sources. Weak data are
noisy, whilst very strong data can often be systematically badly measured
(especially in DENZO, which assumes that all your spots are the same size
and shape) or overloaded. E1 and E2 thus tend to dominate the error model
at high and low resolutions, respectively.
Hope that helps,
Joe



 Does anyone know of a detailed rigorous discussion of how the
 scalepack error model/Bayesian reasoning works? The scalepack manual
 has no equations for this.

 Richard Gillilan
 MacCHESS



Re: [ccp4bb] MAD wavelength

2009-07-17 Thread James Holton

Use the scan.

And yes, don't fry your crystal doing it. The best way to do this is put 
the fluorescence detector as close to the sample as you can and then 
optimize the count rate by attenuating the beam. However, some beamlines 
are set up to hit your crystal with full beam and then back off the 
fluorescence detector to the other side of the room where it can bask in 
the warm sunny glow of your precious crystal. On beamlines like this, 
you must use a sacrificial crystal to do your absorption scan. 
Obviously, I am not a fan of the latter method, but it really does 
depend on what beamline you are going to use. Ask your beamline 
scientist about this!


Note: technically, a MAD scan is not a fluorescence scan, it is an 
absorption scan. You are measuring absorption and using fluorescence as 
a tally. A fluorescence scan would have the fluorescent photon energy on 
the x-axis.


Unfortunately, methods of doing absorption scans are not the only thing 
that differs from beamline to beamline and wavelength calibration is 
notoriously difficult to do with an accuracy of five decimal places (1 
eV in 12 keV). Personally, I calibrate on copper metal, since it is very 
stable and well understood, and the absolute Bragg angle I see from 
silicon powder gives me a consistent wavelength. Some beamlines 
calibrate on SeMet, but the problem with SeMet is that the actual energy 
of the edge is NOT the canonical Se edge found in the literature (it 
is about 1 eV higher). SeMet is also relatively toxic, expensive and 
tends to oxidize, shifting the peak. Probably the best example of the 
shifting edge of Se I know of is figure 1 in Sarret et. al. (2005) 
Applied and Environmental Microbiology, 71(5), 2331–2337. This paper 
contains spectra for nine Se-containing reference compound, a 
substantial fraction of the total that are known.


Because of all this, I have been pushing the use of dandruff shampoo as 
an internationally recognized, inexpensive and particularly rad-hard 
selenium reference. Selsun Blue(R) (and the extra strength variety of 
Head  Shoulders(R)) is 1% selenium sulfide, and gives a nice clear 
spectrum with a sharp white line. What remains is to calibrate this 
stuff against some NIST-traceable Standard Reference Material.


So, my advice is to ask your beamline scientist, and don't forget to 
pack shampoo.


-James Holton
MAD Scientist


Jerry McCully wrote:

Dear All:

Next week we are going to try some seleno-Met labeled crystals.

We checked the literature to try to find out the peak wavelength that 
has been used for SAD or MAD data collection. But they are slightly 
different ( may be 50 ev) in different papers.


I guess this is due to the discrepancy between the fluorescence 
scanning and the theoretical vaules of f' and f''.


When we collect the data, which wavelength should we use? Should we 
trust the scanning results?


Thanks a lot for your comments.

All the best,

Jerry McCully




Insert movie times and more without leaving Hotmail®. See how. 
http://windowslive.com/Tutorial/Hotmail/QuickAdd?ocid=TXT_TAGLM_WL_HM_Tutorial_QuickAdd_062009


[ccp4bb] FPLC price

2009-07-17 Thread Natalie Roy
Hi everyone,

I forgot to mention the price in the advertisement for the FPLC.  We are asking 
$11,000 CAD for the system.  Thanks,

Natalie Roy


[ccp4bb] FW: EPMR statistics

2009-07-17 Thread Jerry McCully









Dear ALL:

  We are trying to do a molecular replacement using EPMR. IT is weird that 
we got high CC(0.58) but with high R factor about 0.89 as well.

The second tier of solutions have cc about 0.54 and R factor about 0.95.

   How can we optimize our EPMR trials to get a definitive correct 
solutions?

   Thanks a lot for the help.

Have a nice weekend!

Jerry McCully

 

Lauren found her dream laptop.  Find the PC that’s right for you.
_
Lauren found her dream laptop. Find the PC that’s right for you.
http://www.microsoft.com/windows/choosepc/?ocid=ftp_val_wl_290

Re: [ccp4bb] Scalepack error model?

2009-07-17 Thread Richard Gillilan

Thanks Joe and others.

Bits and pieces of this story appear in 11.4.8 of International  
Tables volume F, Borek et. al. Acta Cryst D59 (2003) and the  
Scalepack manual, but none are complete or have enough detail to  
follow easily. None of them give the expression for Chi-square for  
this problem.


I found a presentation by Jay Ponder online (for his Bio5325 course)  
that gives:


chi-2 = 1/N sum (I_avg - I_meas)^2/(sigma_avg^2 + sigma_meas^2)

where the sum probably runs over all reflections and the I_avg is the  
average of the appropriate group of symmetry-related reflections.  
Sigma_avg^2 should be the sigma computed from the error model below  
(not given in the presentation) I think and sigma_meas is the sigma^2  
from the actual symmetry-related reflections.


One would then adjust the error parameters below to give chi-square  
approx unity and this leads to the proper scaling factors for  
intensities and sigmas.


One confusing hitch seems to be that (according to the International  
Tables F Eqs.(11.4.8.5) and (11.4.8.6)), the error model is also  
implicitly defined and must be solved iteratively ... though it's  
hard to see that from the text.


Does this sound right?


Richard



On Jul 17, 2009, at 4:12 PM, cockb...@pasteur.fr wrote:


Dear Richard,
I *think* it works like this, don't know if it's detailed or rigourous
enough for you!

If I(l,h,i) is the intensity of the ith observation of reflection h on
frame l, with error sig(l,h,i), and S(l) is the (inverse) scale  
factor to

be applied to frame l, then the error in the scaled intensity
I(l,h,l)/S(l) is parameterised in terms of the error scale factor  
(E1) and

estimated error (E2) as

sqrt( (E1*sig(l,h,i))**2  +  (E2*I(h)*S(l))**2  )

where I(h) is the weighted mean of the scaled intensity values for  
index h

(i.e. the merged, scaled intensity).

The reasoning behind this that the errors in the intensities of  
strong and

weak reflections generally arise from different sources. Weak data are
noisy, whilst very strong data can often be systematically badly  
measured,
especially in DENZO, which assumes that all your spots are the same  
size
and shape, or overloaded. E1 and E2 thus tend to dominate the error  
model

at high and low resolutions, respectively.
Hope that helps,
Joe



Does anyone know of a detailed rigorous discussion of how the
scalepack error model/Bayesian reasoning works? The scalepack manual
has no equations for this.

Richard Gillilan
MacCHESS






Re: [ccp4bb] Scalepack error model?

2009-07-17 Thread Zbyszek Otwinowski

The scalepack log file gives the formula:

 Summary of reflections intensities and R-factors by intensity bins
 R linear = SUM ( ABS(I - I)) / SUM (I)
 R square = SUM ( (I - I) ** 2) / SUM (I ** 2)
 Chi**2   = SUM ( (I - I) ** 2) / (Error ** 2 * N / (N-1) ) )

which equivalent to the Jay Ponder's formula, with the important addition, that 
sigma_avg and Iavg represent the average of all _other_ measurements with the 
same reduced hkl index.

All sigmas are calculated from the error model described in the publications.

Some of the error model parameters are defined at the moment by user, they can 
be refined iteratively by experimenter by adjusting parameters in subsequent 
runs of scalepack, but most of the time it is not required. New version will 
adjust all these parameters automatically.


Zbyszek Otwinowski

Richard Gillilan wrote:

Thanks Joe and others.

Bits and pieces of this story appear in 11.4.8 of International Tables 
volume F, Borek et. al. Acta Cryst D59 (2003) and the Scalepack manual, 
but none are complete or have enough detail to follow easily. None of 
them give the expression for Chi-square for this problem.


I found a presentation by Jay Ponder online (for his Bio5325 course) 
that gives:


chi-2 = 1/N sum (I_avg - I_meas)^2/(sigma_avg^2 + sigma_meas^2)

where the sum probably runs over all reflections and the I_avg is the 
average of the appropriate group of symmetry-related reflections. 
Sigma_avg^2 should be the sigma computed from the error model below (not 
given in the presentation) I think and sigma_meas is the sigma^2 from 
the actual symmetry-related reflections.


One would then adjust the error parameters below to give chi-square 
approx unity and this leads to the proper scaling factors for 
intensities and sigmas.


One confusing hitch seems to be that (according to the International 
Tables F Eqs.(11.4.8.5) and (11.4.8.6)), the error model is also 
implicitly defined and must be solved iteratively ... though it's hard 
to see that from the text.


Does this sound right?


Richard



--
Zbyszek Otwinowski
UT Southwestern Medical Center  
5323 Harry Hines Blvd., Dallas, TX 75390-8816
(214) 645 6385 (phone) (214) 645 6353 (fax)
zbys...@work.swmed.edu


Re: [ccp4bb] EPMR statistics

2009-07-17 Thread Jerry McCully


Here are the solutions from 100 runs(sorry for the long list). Would you please 
help to have a look?

Thanks a million.---Jerry 


Soln  1  Rot= 162.63  98.47 303.11 Trn=  26.48  73.35 105.32 CC= 0.452 R= 0.996
Soln  1  Rot= 111.26 104.75  41.63 Trn=  31.00  24.66  69.53 CC= 0.528 R= 0.961
Soln  2  Rot= 157.54  79.66 260.98 Trn= -14.86  57.85 124.36 CC= 0.513 R= 0.971
Soln  3  Rot= 113.99  91.61 264.87 Trn=   8.06  72.05  53.61 CC= 0.521 R= 0.946
Soln  4  Rot= 337.86  88.06 347.72 Trn=   4.17  75.16 127.45 CC= 0.520 R= 0.952
Soln  5  Rot= 146.81  78.67  75.86 Trn=  56.22  44.05  11.76 CC= 0.546 R= 0.941
Soln  6  Rot= 342.38  80.29 240.59 Trn=  73.71   6.92  85.48 CC= 0.582 R= 0.898
Soln  7  Rot= 321.28 101.88 329.83 Trn=  54.31  64.44  33.42 CC= 0.522 R= 0.967
Soln  8  Rot= 145.15  68.35  63.18 Trn=  73.90  11.90  91.09 CC= 0.545 R= 0.954
Soln  9  Rot= 162.94 101.78 301.17 Trn= -20.34  47.14   1.30 CC= 0.584 R= 0.892
Soln 10  Rot= 166.24 100.33 183.82 Trn=   5.93  64.49  55.04 CC= 0.537 R= 0.957
Soln 11  Rot= 333.36  94.38 303.46 Trn=  63.94  20.82 127.15 CC= 0.519 R= 0.953
Soln 12  Rot= 175.81  92.62 162.15 Trn=  48.20  37.76  12.37 CC= 0.514 R= 0.946
Soln 13  Rot= 359.03  57.14 234.96 Trn= -20.66  77.65  47.22 CC= 0.530 R= 0.949
Soln 14  Rot= 167.05 107.92 178.90 Trn= -32.85  65.33  96.19 CC= 0.545 R= 0.966
Soln 15  Rot= 350.98  29.83  44.82 Trn=  -8.10  37.94 120.78 CC= 0.524 R= 0.957
Soln 16  Rot= 184.84 105.40  18.14 Trn=  67.66  39.49  14.13 CC= 0.509 R= 0.968
Soln 17  Rot= 304.53  52.96 238.53 Trn=  80.59   0.16  92.24 CC= 0.497 R= 0.967
Soln 18  Rot= 339.71  89.20 120.35 Trn=  64.77  13.44  83.29 CC= 0.514 R= 0.952
Soln 19  Rot= 166.17  97.98 319.60 Trn=  -4.93  26.33 113.33 CC= 0.500 R= 0.972
Soln 20  Rot= 177.36  85.10 285.88 Trn=  56.63  20.33  12.66 CC= 0.520 R= 0.951
Soln 21  Rot= 336.13  92.27 305.89 Trn=  17.47   5.71  76.03 CC= 0.517 R= 0.951
Soln 22  Rot= 355.98  68.54 217.40 Trn= -23.88  47.54 120.73 CC= 0.529 R= 0.978
Soln 23  Rot= 291.68  75.51 137.91 Trn=  78.28   2.43 104.53 CC= 0.528 R= 0.960
Soln 24  Rot= 174.44 110.17 202.14 Trn=  52.50   3.31   9.91 CC= 0.534 R= 0.963
Soln 25  Rot= 162.68 100.62  59.90 Trn=   4.23  75.14   1.37 CC= 0.584 R= 0.894
Soln 26  Rot= 312.97 156.01 304.00 Trn=   7.59  32.89 124.82 CC= 0.498 R= 0.979
Soln 27  Rot= 337.74  94.26  42.32 Trn=   3.05  38.15 119.52 CC= 0.516 R= 0.960
Soln 28  Rot= 348.35  72.27 121.56 Trn=  55.07   8.17  77.60 CC= 0.542 R= 0.956
Soln 29  Rot= 343.30  78.23 358.20 Trn=  16.41  14.08  85.72 CC= 0.584 R= 0.890
Soln 30  Rot= 164.24 100.03  60.71 Trn=  51.06  48.06  88.46 CC= 0.581 R= 0.893
Soln 31  Rot= 313.33  42.12 123.12 Trn=   7.97  32.68  38.06 CC= 0.510 R= 0.974
Soln 32  Rot= 343.59  77.50 359.01 Trn=  16.54  68.19  42.23 CC= 0.584 R= 0.891
Soln 33  Rot= 342.42  80.30 240.46 Trn=  26.75  33.91 129.10 CC= 0.583 R= 0.895
Soln 34  Rot= 156.22  80.14 183.92 Trn=  54.43  41.33  11.32 CC= 0.536 R= 0.954
Soln 35  Rot= 159.62 137.49 132.22 Trn=  36.65  77.76  42.97 CC= 0.547 R= 0.946
Soln 36  Rot= 133.66  96.07 124.31 Trn=  32.78  23.18  11.54 CC= 0.521 R= 0.953
Soln 37  Rot= 300.33  86.94 286.01 Trn=   5.94  38.56 119.49 CC= 0.523 R= 0.950
Soln 38  Rot= 348.44  97.82 286.87 Trn=  65.97  43.64   1.05 CC= 0.524 R= 0.969
Soln 39  Rot= 137.08  80.87 328.12 Trn=  57.48  22.76  72.93 CC= 0.526 R= 0.947
Soln 40  Rot= 325.23 108.71 350.80 Trn=  -9.07  27.08  45.15 CC= 0.534 R= 0.976
Soln 41  Rot= 164.88 113.78  48.57 Trn=   7.30  17.76  97.19 CC= 0.534 R= 0.954
Soln 42  Rot= 163.27 101.40  61.88 Trn=   4.00  21.22  44.79 CC= 0.584 R= 0.889
Soln 43  Rot= 292.62 133.66 249.87 Trn=  -7.95  60.75 120.37 CC= 0.521 R= 0.940
Soln 44  Rot= 129.10 101.25  11.47 Trn=  19.75  46.82   0.67 CC= 0.529 R= 0.955
Soln 45  Rot= 333.46  97.93 119.71 Trn=  30.35  25.72   5.35 CC= 0.532 R= 0.938
Soln 46  Rot= 309.89  70.85 232.67 Trn=  15.98  34.33  69.16 CC= 0.500 R= 0.977
Soln 47  Rot= 347.87  92.61  39.46 Trn= -23.72  66.16 111.17 CC= 0.495 R= 0.990
Soln 48  Rot= 339.04  43.16  48.12 Trn= -10.05  57.51  43.93 CC= 0.547 R= 0.944
Soln 49  Rot= 333.86 146.88 262.49 Trn=   7.25  66.77   0.28 CC= 0.517 R= 0.942
Soln 50  Rot= 342.60  79.23 119.95 Trn=  50.97  60.27  85.66 CC= 0.584 R= 0.892
Soln 51  Rot= 149.45  80.28  75.52 Trn=   9.38  17.13  98.47 CC= 0.540 R= 0.945
Soln 52  Rot= 125.50 117.08 129.59 Trn=   7.97  27.45  48.21 CC= 0.509 R= 0.979
Soln 53  Rot= 162.42  72.13   6.02 Trn= -25.09  63.86  34.82 CC= 0.512 R= 0.957
Soln 54  Rot= 288.85  97.23 302.92 Trn=  32.91  21.23  19.83 CC= 0.527 R= 0.947
Soln 55  Rot= 300.72  87.57 167.67 Trn=  10.58  67.02 119.45 CC= 0.522 R= 0.954
Soln 56  Rot= 343.38  77.62 238.67 Trn=  73.47   7.48  85.75 CC= 0.584 R= 0.890
Soln 57  Rot= 345.98  65.40 118.41 Trn=  -9.20  25.75  46.26 CC= 0.511 R= 0.960
Soln 58  Rot= 352.52  87.64 206.73 Trn=  65.26  37.12 127.62 CC= 0.505 R= 0.974
Soln 59  Rot= 295.75  81.01 144.81 Trn=  68.54  38.80   0.98 CC= 0.532 R= 0.954
Soln 60  Rot= 134.83 116.31  94.84 Trn=  66.49  36.90 124.94 CC= 

[ccp4bb] Testers needed for RasMol 2.7.5

2009-07-17 Thread Herbert J. Bernstein

Testers would be appreciated for the release candidate binary kits for the
RasMol 2.7.5 on sourceforge:

http://downloads.sf.net/openrasmol/RasWin_2_7_5_Install_17Jul09.exe
http://downloads.sf.net/openrasmol/RasMol_2_7_5_i686_Slackware_17Jul09.tar.gz
http://downloads.sf.net/openrasmol/RasMol_2_7_5_i386_OSX_17Jul09.tar.gz

The source tarball is

http://downloads.sf.net/openrasmol/rasmol-2.7.5-17Jul09.tar.gz

The manual is available at

http://www.bernstein-plus-sons.com/software/RasMol_2.7.5/doc/rasmol.html

Further documentation and web pages are in progress.l

The MS Windows installer RasWin_2_7_5_Install_17Jul09.exe is run by 
double-clicking it after
download.  The Slackware and Intel Mac OSX installers are run by 
downloading, unpacking,

entering the directory
and executing

rasmol_install.sh --compilefonts

The major changes in this release are support for approximation to
Lee-Richards surfaces by use of the commands
  map generate LRsurf mesh
  map generate LRsurf surface
coloring of maps by
  map color atom
and selection of atoms near a map contour by
  map select atom

as well as preliminary support for the SBEVSL record movie making
commands

  http://sbevsl.wiki.sourceforge.net/Movie+Making+Commands

Bug reports, comments, correction and suggestions would be appreciated

Please report problems to

  y...@bernstein-plus-sons.com

=
 Herbert J. Bernstein, Professor of Computer Science
   Dowling College, Kramer Science Center, KSC 121
Idle Hour Blvd, Oakdale, NY, 11769

 +1-631-244-3035
 y...@dowling.edu
=