Thursday, 6 October 2011

not so good news (Steve Jobs RIP)

From: William G. Scott
Date: 6 October 2011 05:13

On Oct 5, 2011, at 5:52 PM, Bosch, Juergen wrote:

> http://www.apple.com/stevejobs/
>
> May innovation continue to lead the future.
>
> Jürgen

I've been quite saddened about this all evening.  Even though we knew it was coming, it is still incredibly sad, especially at his age (only 56).

Although I never met him, through OS X, he made a fairly significant impact upon my life and career development (such as it is).

The advent of OS X has had a significant impact upon both my professional and personal lives. Although I have worked with unix systems since the mid 1980s, it was only when OS X came along, and I made the switch (from SGI's Irix, which many of us remember fondly as a total pig of an OS), did I really come to enjoy using computers. As a crystallographer, a unix operating system really is essential I think, and when OS X appeared, a fundamental change took place in my office as a young assistant professor: I only needed one computer.

Before, I had an enormous SGI sitting on a table, for doing science. With its stereographic hardware and dials, it put me back (well, actually, the taxpayer) about $12K for an R10,000, which had a 150MHz RISC processor. The sony trinatron monitor did a good job heating the room, especially in the summer. Then I had this little colored hunk of plastic, the first generation bottom-of-the-line iMac, upon which I wrote my papers and grant proposals, sitting on my desk.  This left very little room in my office to mope and feel sorry for myself, which my wife tells me are the only things I am good at.

OS X came on an extra CD packaged with an iBook I bought just before the Sept 11th attack. I remember, because I gave my first presentation using OS X 10.0.3 or something like that on that day, in Grenada, Spain, right next to the Alhambra. It was the worst talk I ever gave, but it wasn't the fault of the software. I was excessively nervous, so much so that my arch-competitor I guess took pity and tried to calm me down. Maybe I sensed something bad was about to happen. Anyway, it was an inauspicious start.

I eventually made it home and put it on that colored piece of plastic in my office and discovered it really was a real unix operating system, albeit in its infancy. I sort of got lost in it for months at a time, and was determined to get all of the software used in macromolecular crystallography running on OS X.  O was about the only thing at the time that seemed ready. What evolved from that was a crystallography on os x website, which started off as a chronicle of my pain, which I seem to feel an almost moral obligation to share and inflict upon everyone around me. (I work at a university that, frankly, isn't very good at infrastructural support, to put it mildly, and I wound up having to learn to do everything the hard way, by myself.)

I think it is fair to say that Steve Jobs and OS X were some of the main things that made this, and my survival, possible.

I've primarily used OS X I guess from the bottom up, but at the same time I was a bit of an Apple fan-boy in the sense that I got the first iPod when it came out, the first iPod Touch, the first MacBook Air, the first iPad, etc. I've never regretted any of those purchases, nor the purchase of the 20 or so Apple computers that I've acquired for work and family over the last decade or so.

My newest OS X adventure has been into computer audio. I really still don't know much of anything about it, but again Apple has made it an enjoyable experience. I bitch and moan about iTunes as much as the next guy, but if you stop to think about it, it transformed the entire music industry, and turned what began as a network of illegal file sharing (Napster), something I regrettably missed out completely, into a legitimate business model. Some might look at this as co-opting, but I think there is a bit more to it than that. In any case, both from the perspective of hardware and software, Apple/Jobs completely changed how I listen to music.

During that same decade I was starting out as an academic scientist, I had more or less given up on music almost completely. My oldest son now likes music (Pearl Jam, Joy Division, REM), but at the time he couldn't handle listening to it at all, probably as a consequence of being somewhere on the autistic spectrum. OS X provided an escape from all that for me, but it eventually also provided a relief for him I think, and through it he came to enjoy (some) music and now plays cello and electric guitar.

So, I guess Steve Jobs has had a large, albeit indirect, influence on my life, and that of my family and research group. I never met him, but I've learned a bit about him through a former VP. I doubt I would have liked him very much personally. Obsessive secrecy and temper tantrums aren't my favorite personality traits. But as I continue to age, I realize superficial likability isn't really that much of an asset.

I'm kind of worried about the future of my favorite platform.

A friend once said to me that Apple and The Gap were the only companies that successfully marketed to my generation (those left to pick up the shattered pieces of the world left to us by the rampages of the infinitely self-entitled baby-boomers). Hopefully Apple doesn't go the way of The Gap.  That is the trouble with a personality cult.  I hope this is something different, and more sustainable.

-- Bill

----------
From: Ashley Buckle
Bill
So eloquently put, much of what you say resonates with me. Very sad day.
Cheers
Ashley

Sent from my iPhone

----------
From: Tim Gruene



On 10/06/2011 06:54 AM, Ashley Buckle wrote:
> [...]
> Sent from my iPhone
 ^^^^^^^^^^^^^^^^^^^^
 How appropriate ;-)
> [...]

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



----------
From: Richard Gillilan
Jobs and Apple not only deeply shaped my career, but also my son's path as well.

I bought an Apple II+ with my first Summer's earnings as an undergraduate lab assistant in organic chemistry in the very early 80's. Spent winter break writing a printer driver in Apple 6502 code so I could plot molecular orbitals.  Skipped the first Macs when they came out, but returned to Apple when the Performa was popular. I may have been the first to announce on this list that OS X would be based on unix ... and I enthusiastically got a copy as soon as I could get my hands on it.

It's true there will never be another Jobs, but I fully expect more great surprises ... the universe is only just getting started.

Richard

----------
From: Prince, D Bryan
I thought I would have nothing to say about this topic, but I am sitting
here at my desk listing to my iPod, remembering fondly my apple IIc
computer I started with in elementary school. I would like to share a
story with you, if you would be kind enough to listen.  I have a son who
is severely developmentally delayed, he is nine, but functions as an
infant due to cerebral palsy and other complications. We were at
Children's Hospital during a recent specialist visit, and sat near
another family who have a child much like mine. This child was playing
with an iPad doing simple spelling games. He, too has been
developmentally delayed, but his parents purchased an iPad for him
because of the interactive touch screen. Then they started with very
simple games like popping soap bubbles. In a year he has gained better
fine motor control in his hands, has learned to communicate better with
his family and has started to truly blossom. I pray that my son will
experience a similar epiphany with his new iPad, but only time will
tell. So, I say thank you, Steve- you will be truly missed.

Bryan




Setup pymol in Mac OS X 10.7

From: Mecy Shi
Date: 21 September 2011 04:04

Dear members,

I want to setup pymol in Mac OS X 10.7, but I didn't do this before.

Who can tell me how to setup and tell me the detail setup procedure.
and if need other programs to setup this.

Thank you very much!

----------
From: Xiaoguang Xue
As I know, you have 3 choices:

1, You can apply a free educational-use-only MacPymol from the company, Schrodinger, LLC. This is the easiest way, just move the MacPymol program into your Program Folder. 

2,  I think you still can try to install Pymol by Fink. Just follow the wiki page of Pymol.(http://www.pymolwiki.org/index.php/MAC_Install).  But there is a compatible problem between Fink and Lion. There is a Lion upgrade notes written by William Scott(http://sage.ucsc.edu/~wgscott/xtal/wiki/index.php/Lion_upgrade_notes). Maybe it can help you.

3, If Fink doesn't work. I think you can try to use MacPorts. It looks like there is no compatible problem between MacPorts and MacOS Lion. The install process is very simple (http://www.macports.org/install.php). I found MacPorts still contain Pymol package, also include apbs plugin. But there is no other protein xtallography software packages, like CCP4 and coot. And It is impossible to install MacPorts and Fink at the same time. So If you also want install CCP4 or coot from source code, it will be very inconvenient. 

Best!

Xiaoguang Xue
--


----------
From: Luca Jovine
...not at all! We have both Fink and MacPorts installed on our machines (in /sw and /opt/local, respectively), and they happily coexist.

HTH, Luca

---------------------------------------------------------------

----------
From: X.Xue
Luca,

Thanks for your correction!

Best wishes!

Xiaoguang


----------
From: Pete Meyer
For a fourth choice, you can do a stand-alone build of PyMOL and its dependencies with relatively little problem (at least on 10.6; haven't upgraded yet).

Pete

Xiaoguang Xue wrote:

----------
From: Petri Kursula
Choice #5: you can purchase a licensed version?
Petri







---

---



off-topic: nonspecific DNA binding


From: Gregory Bowman
Date: 21 September 2011 14:43



Can anyone give examples of proteins that bind to DNA/RNA non-specifically? In particular I'm interested in examples where interactions are limited to the phosphate backbone, and not with bases. If you could email me off-list, I can compile a summary for others that may be interested.

Thanks!
Greg

xplot84

From: Tommi Kajander
Date: 21 September 2011 10:09



HI,  looks like - at least for us on Mac OSX (both snow leopard and Lion), in the latest CCP4 (and already previous versions of CCP4) xplot84 is broken,
the .plt plots dont work (ie programs that produce these are useless for that purpose...). ...worth fixing perhaps?

Tommi

map file specification

From: Pascal
Date: 15 September 2011 08:30


Hi,

I am looking at the specifications of the ccp4 map file format and I am
confused with the number of columns and the number of intervals.

I assume that the number of columns is the grid size but what is the
number of intervals (elements 8-9 in the header)?

Regards,
Pascal

----------
From: David Waterman


Hi Pascal,

The map data is a three dimensional array with dimensions [NC, NR, NS]. On its own, this gives you no information about the grid pitch in the three (crystallographic, not Cartesian) directions, which is determined in fractional coordinates by the number of intervals. That is, along the X direction the sampling interval has size 1/NX. Of course, you need the X, Y and Z lengths (elements 11-13) to convert the fractional coordinate sizes to real space units. There is a further conversion to take into account too: the correspondence between X, Y, Z and C, R, S is not fixed by the format, but is file dependent and described by MAPC, MAPR and MAPS (elements 11-13). 

Hope this helps,
-- David

----------
From: David Waterman


typo: MAPC, MAPR and MAPS are elements 17-19 of the header, but you can see that anyway from the specification.
Cheers

-- David

----------
From: Pascal


Thanks for all the answers. I can almost write a map file.

Only one problem remain: what is the machine stamp (element 54)?
What should I put there? A small fortran program is used to write the file.

Pascal

----------
From: David Waterman


Hi Pascal,

This document makes an attempt at an explanation, but in a somewhat obtuse way: http://conventions.cnb.uam.es/References/maplib.html. The important point is that the machine stamp is 4 bytes that are specific for the architecture that wrote the file. For little endian hardware the stamp is 0x44, 0x41, 0x00, 0x00 while the big endian stamp is 0x11, 0x11, 0x00, 0x00.

Cheers

-- David


Hexagonal myoglobin protocol


From: James Hall

Dear CCP4BB,

I need to grow some myoglobin crystals for testing and ideally they need to be hexagonal. I was wondering if anyone had a protocol which gives these reliably?

Thanks in advance for any replies

James Hall
PhD Student
University of Reading

OIST-CCP4 workshop in Okinawa, 5-9 December 2011

Dear Colleagues,

we are pleased to announce the first  CCP4 structure solution school at the Okinawa Institute of Science and Technology (OIST), Okinawa. All details can be found at http://www.ccp4.ac.uk/schools/OIST-2011/index.php

Title:
"CCP4 school: From data processing to structure refinement and beyond"
Dates: December 5 to 9.
Site: Okinawa Institute of Science and Technology, Okinawa, Japan

The school content:

Software workshop: The rest of the time after data collection will feature many modern crystallographic software packages taught by authors and other experts. It will be organized in three Sections - lectures, tutorials and hands-on trouble-shooting.
There will be model data sets available for tutorials but data, provided by participants, will have higher priority for the hands-on sessions.

Applicants:

Graduate students, postdoctoral researchers and young scientists at the assistant professor level are encouraged to apply. Only 20 applicants will be selected for participation. Participants of the workshop are strongly encouraged to bring their own problem data sets so the problems can be addressed during data collection workshop and/or hands-on sessions.

Application:

Application deadline is end October. Application form, the program, contact info and other details can be found at http://www.ccp4.ac.uk/schools/OIST-2011/index.php

Fees:

There is no fee for the workshop. The students will be responsible for their transportation costs outside Okinawa.  Lodging will be provided at the OIST Guest House "Seaside House".  The workshop will also cover the expenses for all meals and refreshments.

Fadel Samatay, Charles Ballard, Garib Murshudov

Postdoc in novel technique development at Diamond Light Source

Dear All

This is rather off topic for the CCP4 BB but I would like to draw people's attention to developments currently taking place at diamond.
We require a postdoctoral scientist to join an exciting and innovative project to develop methods associated with the soft X-ray imaging of vitrified, intact biological cells. The successful scientist will join a team of people involved in the development of the techniques associated with the soft X-ray cryo transmission X-ray microscope (cryo-TXM) for biology beamline B24, a beamline currently in the design and construction phase at Diamond Light Source. As part of the Beamline B24 project a standalone soft X-ray microscope capable of imaging cells to a resolution of 50nm which will ultimately become the beamline end station has been purchased in order to enable the work to progress in advance of beamline completion in 2014.
If interested please feel free to contact me to find out more!
Additionally, if anyone is interested in the potential scientific impact this new technique might have in structural biology then please get in touch as well!!
All the best,
Liz
Dr. Liz Duke
Principal Beamline Scientist
Diamond Light Source



Wednesday, 5 October 2011

Mac OSX 10.7 Lion

From: Phil Evans
Is there any opinion or experience about whether Lion is ready for crystallographic use? Should I "upgrade"?

Phil

----------
From: Jason Vertrees
Phil,
MacPyMOL works fine on Lion.

Cheers,

-- Jason

--
Jason Vertrees, PhD
PyMOL Product Manager
Schrodinger, LLC


----------
From: William Scott
Hi Phil:

I've found few, if any advantages.  I fear for the future.

I've had problems getting coot to run stereo due to the X11 implementation in 10.7.  Apart from that, no major problems with crystallographic software.

Lion greedily uses memory, and any computer I have with less than 4 gig of memory has become extremely sluggish as a consequence of the "upgrade."  Ideally, you need 8 gig.

Even with that, on my 2010 mini that I use for music playback, I regressed to 10.6.8, because of the audio interface. (It seems less robust, more prone to dropouts and now lacks integer mode output).

Sara has been screaming at me for the last two weeks (nothing us usual in of itself) because Apple decided to get rid of "Save As".

Xcode and the compiler set is free again on 10.7.

I've put some suggestions here for how to get rid of the most annoying new features:
http://sage.ucsc.edu/~wgscott/xtal/wiki/index.php/Lion_upgrade_notes

All the best,

Bill
William G. Scott

Contact info:
http://chemistry.ucsc.edu/~wgscott/

----------
From: Jacob Keller
I have noticed, in new versions of OSes, that there generally is
rampant violation of the concept of "if it ain't broken, don't fix."
Shouldn't there be more moments of delight, when you see they have
solved a previous poorly-engineered feature with an elegant solution?
But a lot of the time, you have to try to figure out how to do what
you did in the last version in this version, albeit for no net gain.
Back to punch cards!

The Friday Curmudgeon
--
*******************************************
Jacob Pearson Keller
Northwestern University
Medical Scientist Training Program
*******************************************

----------
From: William Scott
I forgot to mention how delightful the spelling auto-"correction"  feature can be.  (It should have read "nothing unusual in and of itself").

That, at least, can be turned off.
----------
From: Patrick Loll
Still doesn't beat my all-time favorite, an early Microsoft spell-checker that changed "diffract" to "defrocked."

----------
From: Frances C. Bernstein
Shoshana Wodak once told me that Word kept suggesting
Shoeshine Kodak as the correct spelling of her name.

I just tried my copy of Word and it seems to have improved.

                 Frances

=====================================================
****                Bernstein + Sons
*   *       Information Systems Consultants
****    5 Brewster Lane, Bellport, NY 11713-2803
*   * ***
**** *            Frances C. Bernstein

=====================================================


----------
From: Jacques-Philippe Colletier

Hi,

Overall, the transition from 10.6 is seemingless, crystallographically-wise.

Of course you need to have the new XCode 4.1 installed, and you should also download new, 10.7-dedicated 64bits gcc/gfortran/g77 bundles from http://hpc.sourceforge.net/
And then, CNS, Phenix, CCP4, etc... will just run perfect.
I had to reinstall Coot -- but that's minor.
(Mac)Pymol also works fine, yet (for some reason) uses a lot of resources even when idle.
As per the Upsalla Soft. Factory programs, you'll have to recompile then using the above mentioned bundles, or get already compiled binaries from Mark Harris.
You'll also need to recompile Gromacs (and fftw3) if you're using it -- but it then works great.

I agree with W. Scott on the fact that the new OS is really greedy in terms of resources.
I surely wont upgrade my other, older mac computers that run just fine on 10.6.
But if you have a new Mac, Lion is is really neat.

Best
Jacques


Le Sep 10, 2011 à 2:09 AM, William Scott a écrit :

----------
From: harry powell
Hi

My two ha'porth.

If you are thinking of "upgrading" your sole Mac software development box to Lion I'd say "don't do it unless you like a lot of pain". Anything built on Snow Leopard should run okay on Lion (my Tiger builds seem okay on 10.4, 10.5, 10.6...), so unless you really have an over-riding need to move to 10.7, I'd hang fire.

If you're only running applications and not developing, you can always shout at the developers if things go wrong.

My plan is to install Lion on a spare bootable disk and see what happens - if all else fails, at least I can ignore the upgrade until Apple release 10.7.1, 10.7.2, etc and fix most of their screw-ups.
Harry
--
Dr Harry Powell,
MRC Laboratory of Molecular Biology,
Hills Road,
Cambridge,
CB2 0QH

----------
From: Jürgen Bosch
One really nice feature though comes with what Bill's wife hates. No worries about saving an edited file in many programs. The background versioning is great no more needs of having my_manuscript_###1.doc :-) one file to handle them all. And you can go back to older versions of course. The only disturbing feature about this integrated auto backup is sometimes your Mac goes idle before you can typebecause it is saving another version. This is particularly annoying when you edit large Keynote files.

Jürgen
......................
Jürgen Bosch
Johns Hopkins Bloomberg School of Public Health
Department of Biochemistry & Molecular Biology
Johns Hopkins Malaria Research Institute
615 North Wolfe Street, W8708
Baltimore, MD 21205

----------
From: Herbert J. Bernstein
Dear Colleagues,

 Lion is an reality all developers have to live with.  While I
agree that it would be a bad idea to update one's primary
development environment to Lion, it does seem a good idea to have
at least one system with sufficient memory, disk and good enough
graphics and the new UI (user interface) to be able to wring out
problems, especially on the UI side.

 So the question is:  Is a new MacBook Pro or new MacBook Air
sufficient for such testing, or is something heftier with a
studio display necessary?

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

             
=====================================================

----------
From: Bosch, Juergen
Regarding MBP versus MBA, one of my graduate students just got a new MBA and her machine is much faster than my MBP 2.8 GHz Intel Core 2 Duo with 4 GB RAM. Both machines are running Lion.

The new MBA is hot, only disadvantage you can't lock it down, there's no option for it.

Jürgen


......................
Jürgen Bosch
Johns Hopkins University






----------
From: Sean Seaver
Dear Herbert,

I've come across quite a few people that are using mac books as their main development computer.  This site ( http://usesthis.com/ ) can be an good way to learn about various setups.  A popular trend seems to be using a mac book along with the apple thunderbolt display for more screen real estate.

Take Care,

Sean Seaver

P212121
http://store.p212121.com/

----------
From: Phil Evans
I use my MBP with external screen, keyboard & mouse all the time. The new ones are fast, & mine should easily cope with Lion

My question about Lion was because

1. on the one hand as far as I can see Bill Scott only builds latest stand-alone Coots (0.7...) for Lion, and these don't work on 10.6 (is this true Bill?)

2. on the other hand, I had a report that ccp4mg doesn't work on Lion (is that true Stuart?)

and I need both of these (and don't fancy building either myself)

Phil (still on 10.6)

----------
From: William G. Scott
Dear Phil (and everyone):

1.  I've now got automated build systems for coot for 10.7.1 and 10.6.8.  I just haven't had a chance to get the 10.6.8 one on line.  I'll try to do this today.  (Also, the last few haven't build due to a change in the code with which the compiler can't cope, but hopefully that will be a thing of the past soon.)

2.  Most things build on 10.6.X will run on 10.7.X.  I haven't opened ccp4mg, but will give it a try after I wake up.


3.  I've had trouble with stereo on coot with 10.7.1, but, oddly, the problem goes away if I have a second monitor.

If you update to 10.7, keep a clone of 10.6 just in case it drives you nuts.  There are all sorts of perverse changes, and (unlike the reversal in scrolling direction) not a lot of over-ride options.

I guess this is a glimpse of the post-Jobs era, albeit one provided from an inconvenient vista.

Peace and joy,

Bill


On Sep 16, 2011, at 6:33 AM, William G. Scott wrote:

> Dear Phil (and everyone):
>
> 1.  I've now got automated build systems for coot for 10.7.1 and 10.6.8.  I just haven't had a chance to get the 10.6.8 one on line.  I'll try to do this today.  (Also, the last few haven't build due to a change in the code with which the compiler can't cope, but hopefully that will be a thing of the past soon.)
>
> 2.  Most things build on 10.6.X will run on 10.7.X.  I haven't opened ccp4mg, but will give it a try after I wake up.
>
>
> 3.  I've had trouble with stereo on coot with 10.7.1, but, oddly, the problem goes away if I have a second monitor.
>
> If you update to 10.7, keep a clone of 10.6 just in case it drives you nuts.  There are all sorts of perverse changes, and (unlike the reversal in scrolling direction) not a lot of over-ride options.
>
> I guess this is a glimpse of the post-Jobs era, albeit one provided from an inconvenient vista.
>
> Peace and joy,
>
> Bill

----------
From: Peter Keller
I guess this is one of them:

How Apple's Lion won't let you trash documents
The operating system for the nanny state?

http://www.reghardware.com/2011/09/07/apple_mac_os_x_lion_the_nanny_os/

One of the reasons why I'm sticking with Snow Lepoard for now....

Regards,
Peter.

--
Peter Keller                              
Global Phasing Ltd.,                    
Sheraton House,
Castle Park,
Cambridge CB3 0AX
United Kingdom

----------
From: Andreas Förster
I've bitched enough about all things Mac, but this one's just too good to pass on (from the article that Peter linked to):

"The fundamental issue here is Lion's assumption that you don't know what you're doing, and it's going to ensure you're protected from cock-ups that, in your ignorance, you may make."

The first half of that sentence has always been my major gripe with Mac OS.  If there just were a Pro version for those that know what they're doing, but the iPadification of computing goes exactly the opposite way.


Andreas        Andreas Förster, Research Associate
       Paul Freemont & Xiaodong Zhang Labs
Department of Biochemistry, Imperial College London
           http://www.msf.bio.ic.ac.uk

----------
From: Richard Gillilan
Listening to Jobs speak recently, I got the distinct impression that the end of the era of general desktop computers "PC's" may be on the horizon. Of course that's iPad sales rhetoric, but it may be that the public moves away from general computers and that surely will have implications for scientific computing.

Richard


----------
From: Adrian Goldman
Of course this has to be the case.  I think most of us walk around with with mobile phones that have memory and processing power more powerful than all of the computers available before 1990.

Some of the things that seem difficult in Lion and are nanny-state-ish such as automatic file versioning (not that I am running Lion yet, mind) - well is it the case that all of the experts on the CCP4BB have _never_ deleted a file they didn't mean to delete?  _Never_ had Word (O, Coot) crash taking with it 6 hours worth of changes that should have been saved but weren't?

If so, I feel humbled because I have made these mistakes and more.

Adrian


----------
From: Peter Keller
There is nothing nanny-state-ish (or indeed new) about automatic file
versioning. VMS (which was the major operating system that CCP4
targetted in the pre-Unix days) had it in the 1980's and I don't recall
anyone moaning about it at the time. As you say, it could be a lifesaver
if you overwrote something by mistake. VMS allowed you to purge old
versions completely though, whereas Lion doesn't: that is the
difference.


UV imaging of crystals

From: Harman, Christine
Date: 15 September 2011 20:08


Hi All,
I was curious if any of you have tried or even know if it is possible to adapt a stereoscope (in my case an Olympus SZX10 model) so as to view protein crystals with UV illumination. Basically, I want a cheap manual version of what a Rock UV Imager does.  I know this is probably a crazy dream.  However, I would greatly appreciate any comments, advice or experience any of you may have.
 
Thanks so much,
Christine
 

----------
From: Robert Sweet


I'm not going to respond to the larger group, but I know one can buy LEDs that emit strongly at 280 nm, which would give tryptophan fluorescence. They're about $200, and one could build or buy a control circuit for not much more.  I think this is about what the commercial tools do.  You'd want front illumination.  You can get the LED with a convex lens on the front, giving a focus about 1" away. At that point the light is dangerous -- don't shine it into your eye from that distance.  You'd want to ask your local safety guys to check it out.

We would use it at the synchrotron with a flash circuit that would be synchronized with a video camera -- I think roughly 20ms would do it.

Let me know if it works.

Bob

=========================================================================
        Robert M. Sweet 


----------
From: Andrew Purkiss-Trew


Molecular Dimension do such an adaptor which fits to existing microscopes.

See <http://www.moleculardimensions.com/shopdisplayproducts.asp?id=121&cat=X%2DtaLight%3Csup%3E%99%3C%2Fsup%3E+100+%2D+UV+for+Microscope+>


----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.

----------
From: Ed Pozharski

Do you by any chance know the price?  I can seemingly "order" it through
the website for the hefty price of $0.00, which is too good to be true.

--
"Hurry up before we all come back to our senses!"
                          Julian, King of Lemurs

----------
From: Frank von Delft


A while ago I was trying to be cheap, so we played around with it quite a bit in the lab.  After rediscovering some of the basics of signal-to-noise and microscope transmission efficiency and that sort of rot, I realised that the commercial systems may not be all that ridiculously overpriced after all.  Not if one wants to be able to say something useful about really really small crystals -- the only ones that really matter in the grand scheme of things (big ones are quick to test; little ones must first be optimized = money+time).

But maybe I was just being incompetent.  Happens.
phx.

----------
From: Jürgen Bosch


I once tested such a commercial system in Seattle about 4 years ago. It did not impress me. In particular the discrimination between salt and protein did not work for about 10 different proteins from which we already had collected data. sure those were small between 10 and 100 micrometer. Excuse was to few tryptophans
So in theory it is nice but a cheaper variant might be to add Gfp to your protein and screen for something green.
Jürgen

......................
Jürgen Bosch
Johns Hopkins Bloomberg School of Public Health
Department of Biochemistry & Molecular Biology
Johns Hopkins Malaria Research Institute
615 North Wolfe Street, W8708


----------
From: Andrew Purkiss-Trew


Sorry, I don't know, I think that a new PI at our institute has ordered one from his new equipment budget. But I don't have a price to hand, I can ask though.

--
Andrew Purkiss-Trew
X-ray Laboratory Manager
London Research Institute
Cancer Research UK

----------
From: Bosch, Juergen

I'm replying here to myself :-)

So in an off-board discussion it turns out that the "microscope" in question was a special emitted light and not a UV microscope. So real UV microscopes might be better for the purpose of detecting real crystals.

Sorry for the confusion - had too much sun today :-)

Jürgen
......................
Jürgen Bosch
Johns Hopkins University







----------
From: Edward A. Berry <BerryE@upstate.edu>
Date: 16 September 2011 03:07
To: CCP4BB@jiscmail.ac.uk


A "real" UV microscope requires quartz optics, right?
Probably conventional microscopes use glass.
And you can't see 280 nm (and its not good for your eyes)
so you need some kind of phosphor screen to view the image?

----------
From: Nagarajan V <tatoosh@gmail.com>
Date: 16 September 2011 04:57
To: CCP4BB@jiscmail.ac.uk


Typically, what you image is Trp fluorescence by exciting at around 280 nm and observing at around 350 nm. Standard silicon based detectors do fine at the detection wavelength, although, as you can imagine, increased sensitivity in the UV means increase in the price of the detector. If your excitation and emission light paths do not overlap, you also can get by with standard glass (crown, flint, etc.) optics since they do allow some of the 350-nm light to get through. Therefore, yes, it is possible to build an inexpensive UV imager based on inexpensive excitation light source (Douglas Instruments offers a pen light), and standard lab microscope. Of course, for increased sensitivity and contrast you need a very good light source, optics made of quartz and calcium fluoride that let almost all the UV light through, highly discriminating filters and a sensitive detector.

V. Nagarajan
JANSi

----------
From: Klaus Fütterer <k.futterer@bham.ac.uk>
Date: 16 September 2011 09:19
To: CCP4BB@jiscmail.ac.uk


From the experience when our (commercial) UV imaging system was set up, I can confirm that signal-to-noise is a non-trivial parameter for imaging in the UV range.

I find the additional info gained from the UV capability very useful, not just to distinguish salt from protein crystals, but also to tell protein from buffer precipitate, buffer phase separation from protein phase separation, etc.

Klaus


=======================================================================

                   Klaus Fütterer, Ph.D.
               Reader in Structural Biology
                 Undergraduate Admissions

School of Biosciences             P: +44-(0)-121-414 5895
University of Birmingham          F: +44-(0)-121-414 5925
Edgbaston                         E: k.futterer@bham.ac.uk
Birmingham, B15 2TT, UK           W: http://tinyurl.com/futterer-lab
=======================================================================

----------
From: Shiva Bhowmik <gene1267@gmail.com>
Date: 16 September 2011 18:13
To: CCP4BB@jiscmail.ac.uk


Would be curious to know the current limitations on UV microscopy employed for screening protein crystals - such as content of aromatic amino acids, protein size etc.

Cheers,

Shiva

----------
From: Patrick Shaw Stewart <patrick@douglas.co.uk>
Date: 19 September 2011 12:35
To: CCP4BB@jiscmail.ac.uk



Hi Christine

As Nagarajan mentioned, we do have a "UV Pen" that we offer at a reasonably modest price.

However we have not been promoting it because we've been struggling to find a camera that we can recommend to use with it.

We use it with an old Russian microscope (normal glass optics) and a cheap consumer Panasonic camera on a "starlight" setting, which works very well.  We developed it for our own crystallization project and we have found it very useful, although we occasionally get both false positives and false negatives.

One problem is that the cheap camera we use is now obsolete, and we are investigating our own solution, which will involve writing our own software for an off-the-shelf CCD and lens.  (CCD seems better than CMOS.)  If anyone knows of a suitable more expensive SLR-type consumer camera, please let me know.  We can't find one.

We are looking at just the tail end of the fluorescence, the part that is in the visible or near-UV spectrum, so you need a sensitive camera.  We have found that you need to push everything, taking e.g. 30 exposures and combining them, subtracting the "dark" image etc.  Presumably the Panasonic point-and-click camera does this.

One solution is to buy a camera from one of the microscope suppliers, but they use very old technology, and charge you a fortune for e.g. the controller (which is essentially a laptop that was designed 15 years ago).  The results are no better than the Panasonic camera, and we hesitate to sell you a UV source for around 1000 pounds then recommend a camera system for 6,000.  However the pens are available to anyone who wants to order one!

We're not claiming that our approach is as good as the more expensive commercial systems, some of which have quartz optics.  Each image takes 30s, you have to work in a dark room, and you can't use it in transmission mode (allowing you to see absorbance) for which you need quartz optics.

Another way to go is to use the approach that we published recently, where you covalently (therefore unambiguously) label your crystals after they have grown.  It's pretty much 100% accurate.  See the methods section of the ref below.

Hope this helps

Best wishes

Patrick


Patrick D. Shaw Stewart, Stefan A. Kolek, Richard A. Briggs, Naomi E. Chayen and Peter F.M. Baldock. "Random Microseeding: A Theoretical and Practical Exploration of Seed Stability and Seeding Techniques for Successful Protein Crystallization"

Crystal Growth and Design, 2011, 11 (8), p3432. 

 

On-line at http://pubs.acs.org/doi/abs/10.1021/cg2001442

 


--
 patrick@douglas.co.uk    Douglas Instruments Ltd.
 Douglas House, East Garston, Hungerford, Berkshire, RG17 7HD, UK
 Directors: Peter Baldock, Patrick Shaw Stewart

 http://www.douglas.co.uk
 Tel: 44 (0) 148-864-9090    US toll-free 1-877-225-2034
 Regd. England 2177994, VAT Reg. GB 480 7371 36



Opportunities in Newcastle

Incase the (extremely attractive) lectureship advertised by Rick is a little rich for your blood, I'd like to draw your attention to postdoctoral positions available in the Endicott and Noble groups, currently being established in the Northern Institute for Cancer Research in the Medical Sciences Division of Newcastle University.

The posts, supported by the MRC, are available for up to (nearly) five years. They represent an opportunity to join a new interdisciplinary team, looking at basic and translational structure/function studies relating to cell cycle regulatory mechanisms.

Newcastle University is investing heavily in developing structural biology, and we are coordinating our activities closely with Rick's team, less than 100m away at CAMB, to achieve a critical mass for crystallography and other structural biology techniques in this area. Frankly, if I were a postdoc, I'd apply.

Informal queries can be addressed to Jane or I, but no application will be considered unless it is made through the University portal.  The posts are advertised on the Newcastle University website with the identifier:

A1014R (NICR)

This is currently accessible through the URL:

http://tinyurl.com/6bcz8rl

Best wishes,

                       Martin

Professor Martin Noble
Northern Institute for Cancer Research
Paul O'Gorman Building
Framlington Place
NE2 4HH



Industry position job posting


Date: 5 October 2011 17:58


Hi All,

The protein engineering department at Zymeworks Inc. (www.zymeworks.com), is actively searching for talented structural biologists who are interested in doing in silco protein engineering.

If interested please visit [http://zymeworks.com/careers/postings.html]

Cheers,

Paula Lario Ph.D.
Zymeworks, Inc.