Antwort: Re: WG: [RASMB] AUC Upgrade

Tue Nov 14 07:30:24 PST 2006

Hallo Peter,

thanks for the hint about using the rmsd in order to identify & eliminate
scans with wrong cell extension. This is indeed what we use when we
evaluate SedVel with a fitting approach, and the corresponding menu item in
sedfit is very helpful !

However, for overlayering experiments and density gradients (which
occasionally we also run on the XLI, normally on Schlieren optics), we
evaluate without fitting, hence without rmsd. The same must apply, I
believe, for M*-evaluation and others. Ordering scans by hand is
painstakingly slow :-(

I still hope - thanks for the update, Mr. Perdue !! - that this bug can be
truely fixed.

Beste Grüße, Wendel

Dr. Wendel Wohlleben
Polymer Physics

Phone: +49 621 60-95339, E-Mail: wendel.wohlleben at basf.com
Postal Address: BASF Aktiengesellschaft, GKP/O - G201, D-67056
Ludwigshafen, Germany

BASF - The Chemical Company

BASF Aktiengesellschaft, Registered Office: 67056 Ludwigshafen, Germany
Companies' Register: Amtsgericht Ludwigshafen, HRB 3000
Board of Executive Directors:
Juergen Hambrecht, Chairman; Eggert Voscherau, Vice Chairman;
Kurt W. Bock, Martin Brudermueller, John Feldmann, Andreas Kreimeyer, Klaus
Peter Loebbe, Stefan Marcinowski, Peter Oakley
Chairman of the Supervisory Board: Juergen Strube

             Peter Schuck                                                  
             <pschuck at helix.ni                                             
             h.gov>                                                     An 
                                        wendel.wohlleben at basf.com          
             14.11.2006 16:08                                        Kopie 
                                        rasmb at server1.bbri.org             
                                                                     Thema 
                                        Re: WG: [RASMB] AUC Upgrade        

Hi Wendel,

we also that intermittently the problem on our newest XL-I that individual
outlier scans seem to have been saved with the wrong extension (wrong cell
number).  However, I was under the impression that it had been fixed
quickly (i.e. within a few months), and we don't seem to have that problem
anymore.  All of that was one or two years ago.

During the time when it did occur, things were not as bad for us as it
sounds, since each of the interference cells has its own 'fingerprint' with

the time-invariant noise.  (The problem did not occur with the absorbance
cells.)  Due to the characteristic noise, the few scans that were saved
incorrectly could be easily flagged out once a preliminary fit was done,
since the rmsd of that particular scan was clearly an outlier due to
different TI noise offset.  There is a function into SEDFIT for recognizing

and excluding the worst scan, which was included just this purpose.  Note
that this approach works only for samples where you can get a fit with rmsd

< 0.01, which is usually the case for the fairly dilute protein systems (<
1 mg/ml) we study.  It would probably not work if you have high loading
concentrations such that the TI signal is masked by residual systematic
deviations in the fit.

Best,
Peter

At 08:52 AM 11/14/2006, you wrote:
>Hallo Christine,
>
>we had both of our XL-I upgraded in february 2006, exactly in the way you
>describe it: PC, CCD, software:
>
>The dark noise of the new CCD and new frame-grabber is somewhat better,
but
>the pixel size and number is identical, leading to no significant
>enhancement in fringe shift accuracy.
>The new software is the old software, with only 1 new feature: Data from
>cell x is 'sometimes' saved as data from cell (x-1). 'Sometimes' means a
>few % of all scans are missorted. The problem is aleatoric, independent of
>waiting time between scans and less dramatic on one of the machines. We
can
>eliminate scans from the analysis, depending on bad residuals in fits, but
>what if you run two very similar samples - how do you identify a missorted
>scan? What I find incredible is that this bug was known with Beckman
before
>our placement of order, but they told us only after the upgrade !!! Since
9
>months they are unable to fix it.
>
>Now, why did I spend the money for virtually no progress? Simply because
my
>IT-department no longer tolerates Win95-PCs in the network. If you can
>still escape ;-) don't upgrade, wait for Tom Laue's next-generation-AUC.
>
>
>Beste Grüße, Wendel
>
>
>Dr. Wendel Wohlleben
>Polymer Physics
>
>
>Phone: +49 621 60-95339, E-Mail: wendel.wohlleben at basf.com
>Postal Address: BASF Aktiengesellschaft, GKP/O - G201, D-67056
>Ludwigshafen, Germany
>
>
>BASF - The Chemical Company
>
>
>BASF Aktiengesellschaft, Registered Office: 67056 Ludwigshafen, Germany
>Companies' Register: Amtsgericht Ludwigshafen, HRB 3000
>Board of Executive Directors:
>Juergen Hambrecht, Chairman; Eggert Voscherau, Vice Chairman;
>Kurt W. Bock, Martin Brudermueller, John Feldmann, Andreas Kreimeyer,
Klaus
>Peter Loebbe, Stefan Marcinowski, Peter Oakley
>Chairman of the Supervisory Board: Juergen Strube
>
>
>
>
>
>             "Christine.Wandre
>             y at epfl.ch"
>             <Christine.Wandre                                         An
>             y at epfl.ch>                rasmb at server1.bbri.org
>             Gesendet von:                                          Kopie
>             rasmb-bounces at ser
>             ver1.bbri.org                                          Thema
>                                        [RASMB] AUC Upgrade
>
>             12.11.2006 18:43
>
>
>
>
>
>
>
>
>Dear all,
>
>Is there anybody who has already upgraded an OPTIMA XL-I??
>
>We purchased our equipment in 1997. It is still
>very well functioning and was not too much used.
>Now I intend replacing the PC, software, digital
>camera, software,...- the complete set offered by
>Beckman.
>
>Is there anything special that I should consider?
>
>Thank you,
>Christine
>--
>PD Dr.habil. Christine Wandrey, MER
>Ecole Polytechnique Fédérale de Lausanne (EPFL)
>Laboratoire de Médecine Régénérative et de Pharmacobiologie (LMRP)
>http://people.epfl.ch/christine.wandrey
>
>tél: (+41) (21) 693 96 61 in office AAB 042 or
>(+41) (21) 693 36 72 in office CH J2 494
>fax: (+41) (21) 693 96 85
>
>postal address
>EPFL-SV-IBI-LMRP
>AAB 042, Station 15
>CH-1015 Lausanne
>
>_______________________________________________
>RASMB mailing list
>RASMB at rasmb.bbri.org
>http://rasmb.bbri.org/mailman/listinfo/rasmb
>
>_______________________________________________
>RASMB mailing list
>RASMB at rasmb.bbri.org
>http://rasmb.bbri.org/mailman/listinfo/rasmb