Uppsala Software Factory - Did you know ?

Uppsala Software Factory - Did you know ?

A collection of honourable and dishonourable mentionings of the Uppsala Software Factory, its President (GJK) and his partner in crime (TAJ) on the Internet and in printed media.

For more serious stuff, read this.

Sometimes people want to help our "factory":

subject: Foreign commerce

realname: Licenciado Martinez

email: serviceconnection2000@yahoo.com

From: Gerard - MailUs

MESSAGE: Dear sir,
As a software developer we thought you might be 
interested in selling your products in Mexico 
city. Please let us know if you'd like to contact 
an international distributor. Please send us 
information.

All the best,
Licenciado Martinez,
Sales representative

DEJAVU ? Well, apparently there's more than one program by that name ... looky here !
Why would Elsevier send us a brochure about a new book about bone biology (here) ? The only explanation I could think of is that they picked up our names from this page and drew the wrong conclusion ...

Here you can read that the USF CEO has finally had a plot named after him:

Q18: Where is the best place to look to see where the model needs improving? 

A: Atoms that are involved in restraint violations, large positive or negative difference electron density,
regions with high and/or very anisotropic displacement parameters, violations of non-crystallographic
symmetry, and residues in disallowed regions of the Ramachandran plot. The .lst file contains most of this
information, and SHELXPRO may be used to plot displacement parameters or NCS differences as a function
of residue number as well as Ramachandran and Kleywegt plots.

"Ohne Worte":
.
The true story of "O":
.
Some people take the USF CEO's .signature too seriously:
.
Others have even more problems with his name:
.

Here you can read about the "Rfree police":

We can avoid overfitting the data by monitoring the cross validation R-factor, which is sometimes called the
free-R.   I won't say much about the theory of cross validation.  I will just point you to a few MUST READ
papers.  MUST READ means you MUST READ these papers or risk being embarrassed during group
meetings or at conferences.  Worse still, you might be accosted by the Rfree police. 
  
     Assessment of phase accuracy by cross validation, by Brunger (1993) Acta Cryst. D49, 24-36. 
     Checking your imagination:  applications of the free R value, by Kleywegt and Jones, (1996) Structure,
     4, 897-904. 
     Model building and refinement practice, by Kleywegt and Jones, (1997) Methods in Enzymology, 277,
     208-230. 
     Free R value:  Cross validation in crystallography, by Brunger (1997) Methods in Enzymology, 277,
     366-396.

The USF software is included in the Linux4Chemistry listing Linux4Chemistry listing.
We came across this ad for the "Essens Specialist" who sells strands for only 10 SEK (while stocks last, of course):
.

Found on the web (here):

Gerard Kleywegt was then warmly received
back to give the second plenary lecture. Unfortunately his intriguing talk was not enough to keep all of the
audience attentive. Gerard decided half way through that Lindsay had perhaps had one glass of wine too many and
had to be awakened with what appeared to be an industrial sized cattle prod. Those who could cope with the strain
of "More informalities" found the bar open until the wee sma' hours.

O is such a popular program in Munich that they even organised a little party in its honour ! See here and here ! Thanks to Alexandru Tudor Constantinescu for submitting these pictures !
Some of our CEO's less than serious contributions to the O-info, CCP4 and X-PLOR bulletin boards have been preserved for posterity (here).

Back in 1997, Dave Schuller posted this (note signature !):

On Thu, 18 Dec 1997, anja rabijns wrote:

> Based on the Matthews coefficient I know that there are four molecules in the 
> asymmetric unit.

You do NOT know!

1) The Mathews coefficient for known protein structures empirically falls
into a broad range which corresponds ~ 40 - 70% solvent. there is no
theoretical limitation, except that the solvent content should not be less
than 0% or greater than 100%.  let me repeat: the distribution found is
EMPIRICAL.  new examples could be found which extend the range.

2) for your example, guessing that you assume a solvent content of ~ 50%
for 4 monomers, then each monomer takes up ~ 12.5% of the asymmetric unit.
i would find anything from 2 monomers (-> 75% solvent) to 5 monomers
(->37.5% solvent) perfectly believable.

I think you should first be clear on what you know vs. what you have
estimated if you wish to learn new things.

=======================================================================
"They know.
 They know who belongs, and they know who does not, and they know how to
 make those who do not belong know that they know."        Peter H. King
=======================================================================
                        Dave Schuller
                        Gerard for a day
                        University of California-Irvine
                        schuller@uci.edu

To which our CEO replied:

Dave Schuller bashed a poor Belgian student:
>
[snip]
>

and signed his lecture:
>                        Dave Schuller
>                        Gerard for a day
actually, i have grown mellow in my old age and sent
her a nice e-mail earlier today explaining about
translational ncs and how she may have two dimers
related by an almost pure translation of (0.2,0,0)

--Gerard
  Mr Niceguy for a day

Since we started monitoring what kind of search-engine queries lead to "hits" on our web site, we have had quite a few surprises (and a lot of laughs !). Some of the queries that have lead people to USF pages:
- "disk crash"
- "mutatis mutandis"
- +Dave +Cook +Karate +Uppsala
- Mickrosoft pacman
- RAVE TUNES-PRISONER CELL BLOCK H
- Vilken sträcka trafikeras av snabbtåget AVE?
- Why is the seaman big
- i need to se a model of the atom calcium
- make halloween mask from scratch
- the banana splits
- toplist hetero
- erotic hetero stories
- www busty dcup ,com/

Recently, the Uppsala Software Factory received the following E-mail which almost tempted us into buying a forklift truck (especially since the prices are apparently very reasonable - they know how to attract Dutch customers, obviously):

Subject: Forklift trucks

Dear Sirs,

We have just started our web presentation and would like to 
announce the products we are producing and can be of interest for you.

We are producing diesel, gas and electric industrial forklift trucks.
For any purpose and for very reasonable prices. Just see us at:

http://www.dimex.evro.net

Please, contact us if you are interested.

Best Regards,
Dinko Georgiev

Some of the spellings of the word "Uppsala" that have been fed to search engines in the hopes of locating the USF web site (and still gave the correct result): Upsalla, Upsala, Upsella.
The Uppsala Software Factory is a "Key Resource" under "Crystallographic Computers" and "Crystallographic Software" at Links²Go.
In February 1998, the Uppsala Software Factory was at the Institute of Molecular Biophysics of Florida State University.

And this is what Lynn Ten Eyck once said in the newsgroup bionet.crystallography:

Subject:      Re: b-factor refinement
From:         "Lynn F. Ten Eyck" 
Date:         1997/03/13
Message-ID:   <3328F0AC.496D@sdsc.edu>
Newsgroups:   bionet.xtallography

Howard Flack wrote:
> 
> Ask Lynn van Eyck.
           ^^^
      Lynn Ten Eyck, actually.

There is an excellent discussion of refinement practice by Kleywegt and
Jones to come out in "Methods in Enzymology", but in the mean time
available at http://xray.bmc.uu.se/gerard/gmrp/gmrp.html -- and I
strongly recommend that you download this while you can.  It doesn't so
much contain explicit cookbook advice on when to refine what as it does
advice on how to tell if what you have done is appropriate for your
data.  This is actually the correct problem to address.  The approach
relies heavily on Rfree.

The same server used to have a copy of the Kleywegt & Jones Structure
article,  Gerard J Kleywegt and T Alwyn Jones: Where freedom is given,
liberties are taken [Ways & means] Structure 1995 3 : 535-540 which has
to be one of the more enjoyable diatribes I have read in a long time. 
Alwyn and Gerard would have done very well as Old Testament prophets.

Dale Tronrud has been discussing some of these issues for a long time;
you might look at the following reference.

TRONRUD, D. E. Knowledge-based B-factor restraints for the refinement of
proteins. J. Appl. Cryst. 29, 100 (1996).

If you are going to refine your B-factors much you really must use the
bulk solvent correction and all of the low resolution data.  This has
the added bonus of making your difference maps significantly clearer.

Lynn Ten Eyck
teneyckl@sdsc.edu

In his review of volumes 276 and 277 of Methods in Enzymology, Jan Drenth (Structure 6, Feb 1998, pp. 243-4) writes:

The contributions of Alwyn Jones (and his co-authors) are
extremely useful for the novice in model building and refinement
practice.  They are also a useful refreshment course for the
experienced crystallographer, although the schoolmaster style
of writing may irritate some readers.

During the 1996 Summer School at the Karolinka Institute, Lennart Nilsson referred to us as "The Uppsala Structure Police".
The best program never written by the Uppsala Software Factory has to be APRIL !

In the IUCr Newsletter (vol. 6, nr. 2, p. 10), Mitch Guss writes about the meeting of the Society of Crystallographers in Australia (held in Queenstown, NZ, April 1997):

The relentless growth of protein crystallography in the region was
heralded by the opening talk of G. Kleywegt who together with his
Swedish colleague A. Jones has set out to inflict some of the
accepted standards of small molecule crystallography onto the world
of protein structure.

In her review of volumes 276 and 277 of Methods in Enzymology, Elinor Adman (Acta Cryst D54, 709-710, 1998) writes:

The chapter on detecting folding motifs, by Kleywegt and Jones, describes
tools for detecting similar folds, even during the process of map fitting, and
includes good references to the larger community of structural
consumers interested in classifying folds and detecting similarities.

(but not a word about the other chapter ...)

Way back in 1994, Laurence Pearl proposed to honour and immortalise Gerard's name as follows:

From:   SMTP%"pearl@bsm.biochemistry.ucl.ac.uk"
To:     "Members of CCP4BB mail list" 
CC:
Subj:   In Honour of Gerard K...................

Resent-From: 
Date: Thu, 25 Aug 1994 09:34:02 GMT
Resent-Date: Thu, 25 Aug 94 10:35:51 UT
Message-ID: <94825103551.~INN-ZQAa00165.ccp4bb@dl.ac.uk>

*******************************************************************************

                           HONOURING GERARD

*******************************************************************************

>From time to time you get a derivative that has a big loud isomorphous
difference, and looks at first sight as though it should really solve the
structure.

However as time goes on it becomes clear that this big derivative will not
deliver the phasing power it first promised, and in the end it contributes
nothing to the final map except some irritating and unnecessary noise.........

In keeping with the fine old crystallographic tradition of naming things 
after people (Bijvoet, Friedel, Laue, Harker, Patterson etc.), I would like
to propose that we honour our dear colleague Gerard (with whom it has so much
in common), by naming this kind of derivative a 'Kleijwegt'.                    
I do hope this nomenclature catches on ......

Laurence Pearl

*******************************************************************************

Latest update at 14 January, 2002.