[RASMB] XL/I Absorbance Problem (follow-up)

John Philo jphilo at mailway.com
Thu Aug 23 09:24:25 PDT 2007


I think a couple of points regarding the use of pseudo-absorbance were not
sufficiently brought out by either Bo Demeler or Peter Schuck:
 
(1) With pseudo-absorbance there are vertical shifts of the scans from one
scan to another, somewhat similar to the TIN or 'jitter' in interference
data. Treating this systematic noise as a true constant vertical
displacement certainly helps, but if you look closely at the structure of
these shifts they do not appear to be truly constant across the cell. One
source of this effect is clearly the adjustment of the photomultiplier
voltage based on the light levels at 6.5 cm, as Peter mentioned, and thus
the shifts tend to be largest as the boundary moves past 6.5 cm.
 
(2) I've seen pseudo-absorbance data gathered by a number of users of DCDT+
and that strongly suggests that this approach really only works well when
the absorbance in both sectors is pretty low. At higher ODs the systematic
errors associated with (1) and other effects appear to get much worse.
Exactly how high an OD is reasonable to use depends on the quality of the
data you need, so it is hard to give a general cut-off, but I would probably
stay below ~0.5 OD. Anyone planning to use pseudoabsorbance should do their
own testing to see what errors are acceptable for their particular
application.
 
With respect to (1) it would be useful for Beckman to provide an option that
suppresses the photomultiplier voltage adjustment after the first scan of
each cell, and that also drops the assumption that the voltage should be
initially optimized for measuring sample - reference.
 
Best regards,
 
John
-----Original Message-----
From: rasmb-bounces at rasmb.bbri.org [mailto:rasmb-bounces at rasmb.bbri.org] On
Behalf Of Peter Schuck
Sent: Thursday, August 23, 2007 5:10 AM
To: rasmb at server1.bbri.org
Subject: Re: [RASMB] XL/I Absorbance Problem (follow-up)



Hi All, 

we've used the intensity data acquisition for a long time now after we
discovered that the noise structure of log(raw intensity) data is similar to
that of interference optical data and can be treated with the same algebraic
noise elimination technique.  However, this also introduces the same
disadvantages that we have in interference optical detection from the lack
of an optical reference baseline.  Therefore, we only use it for doubling
the rotor capacity if necessary.  One should mention one particular downside
of the use of different samples in the two sectors, which is that one can't
have a high absorbance in the reference sector at 6.5 cm, because that's a
reference point for the photomultiplier voltage adjustment.  More details
are in the original publication Anal. Biochem (2000) 285:135-142
http://www.sciencedirect.com/science?_ob=ArticleURL
<http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6W9V-45FK4WR-4X&_
user=10&_coverDate=10%2F01%2F2000&_rdoc=1&_fmt=&_orig=search&_sort=d&view=c&
_acct=C000050221&_version=1&_urlVersion=0&_userid=10&md5=9f4421898d4498cb524
5fdeaf9931f41>
&_udi=B6W9V-45FK4WR-4X&_user=10&_coverDate=10%2F01%2F2000&_rdoc=1&_fmt=&_ori
g=search&_sort=d&view=c&_acct=C000050221&_version=1&_urlVersion=0&_userid=10
&md5=9f4421898d4498cb5245fdeaf9931f41

So far I did not notice a marked improvement of signal/noise ratio when
using pseudo-absorbance instead of absorbance data.  Marc Lewis has
rigorously analyzed in detail the statistical properties of intensity versus
absorbance data (see Methods Enzymology 2004, 384:232-242).  As I see it,
the problem is more that introduced by taking the logarithm of the
photomultiplier counts, which causes nonlinearities in the error
propagation, but I think this is more of an issue in sedimentation
equilibrium analysis, where the question of statistical weights for the data
points can have a higher impact.  

If you use TI noise subtraction from the regular absorbance data, you can
get excellent signal/noise ratio, as well.  I'm puzzled by the 0.3 OD
example, which is actually well in the range of OD values that can be very
conveniently studied by regular absorbance, usually with or without TI
noise.  In fact, velocity data down to 0.03 or less absorbance units can be
quite fine for c(s) analysis in SEDFIT, as repeatedly shown over the years
by a number of different labs.

Peter 



Borries Demeler wrote: 

Although many people record the lamp wavelength/intensity profile from

time to time e.g. to check wavelength calibration, I would recommend

also taking every so often a radial scan (collect intensity data) on an

empty hole at your favourite wavelength. This will rapidly show up any

developing problems in the lamp/monochromator/PMT system and help to

distinguish them from cell-associated problems. I'm quite happy to

compare "before and after" scans with anyone who thinks they have a

similar problem, just email me.

    



Hi Andrew,

I think this is very good advice. In fact, the majority of our velocity

absorbance data are now actually collected in intensity mode. For

velocity data, time invariant noise that will always show up much more

dramatically in intensity mode (when no reference data is subtracted)

can be used for determining if the machine is working properly in the

way Andrew suggested and simultaneously can be cleaned up by TI noise

removal procedures (in ultrascan while doing a 2-dimensional spectrum

analysis) so the data can actually be used.



I would like to add one more point:



The results you can get from the intensity acquisition are remarkably

good (and after TI noise removal actually better than absorbance data)

because the stochastic noise from the reference data acquisition is not

convoluted with the stochastic noise that was acquired with the sample

channel. This buys an approximate square-root of 2 factor improvement

in random noise. As a result, you can measure at lower ODs and still

get acceptable data without too much noise. Another benefit is that

you can load 2 different samples in each cell and double your capacity.

I also recommend to do a water scan to get an intensity variation trace

that can be used to get improved I_0 values for the pseudo-absorbance

conversion. A routine is included in UltraScan to help with that.



I am attaching an example of a BSA run with just 0.3 OD absorbance before

and after TI noise subtraction using the 2-dimensional spectrum analysis

in UltraScan. You can see that even at 0.3 OD the data have remarkably

small random noise contributions and are quite suitable for routine

analysis after TI noise subtraction. The later versions of UltraScan now

also include modules for pre-processing intensity data and converting

it to pseudo-absorbance data.



Regards, -borries

  


  _____  


_______________________________________________

RASMB mailing list

RASMB at rasmb.bbri.org

http://rasmb.bbri.org/mailman/listinfo/rasmb

  

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://list.rasmb.org/pipermail/rasmb-rasmb.org/attachments/20070823/69c05a57/attachment.htm>


More information about the RASMB mailing list