[RASMB] standard for sv

[Arthur Rowe]

Hi Marcello -

The question you ask is a good one - although not easily answered. I have a
paper currently submitted on this general area (absolute s values, what
questions can be answered by modelling, and so on). I will try and summarise
the points most relevant to your question:

(1) there is no real point including a 'standard' in your experiments,
except as noted below, point (3). All you will do is compound the errors
(your experiment plus the 'standardisation' experiment). PSL spheres are
tempting as standard particles, but their density is so close to that of a
typical aqueous solvent that errors in their vbar become very serious. Also,
the SE of the estimate of their mean diameter may look encouraging - but not
all that encouraging when you think how critical this parameter is when you
are trying to predict absolute friction (which is what is at stake), not to
mention their absolute mass (oh dear - errors in radius x 3!)

(2) If you are attempting to determine 'absolute' s values, then the biggest
errors come in the temperature uncertainty. There is good stuff in the RASMB
archives about this. At the end of the day, the accuracy of the XL-I/A
temperature system is ±0.5 degrees only - not as good as one would like. But
the precision is better than this, and one can up the accuracy by using the
Liu & Stafford (1995) procedure.

(3) HYDRO will only predict an absolute s value if one knows the 'hydration'
term (aka the 3rd term in the scaled particle theory equation). One never
knows this very well - but there is a decent way round this. Provided one
has a reference conformer - i.e. a particle of the same (or extremely
similar) nature to one's test system (a protein-DNA complex in your case),
then one can assume that the 'hydration' term cancels if one measures all
one's s values relative to the reference conformer*. That means running
(reference conformer) and test species) in the same rotor. It's even better
if one does 'difference sedimentation' (see RASMB archives again). Relative
s values - for these we are talking down to 0.01% precision,  enabling one
to test a range of possible models rather critically.

*having pushed one's reference conformer's details through HYDRO, of course

Hope these thoughts are of some help

Arthur

Reference

Liu, S, Stafford, W  F (1995) An Optical Thermometer for Direct Measurement
of Cell Temperature in the Beckman Instruments XL-A Analytical
Ultracentrifuge  Analyt Biochem  224: 199-202.

--
*******************************************************
Arthur J Rowe
Professor of Biomolecular Technology
NCMH Business Centre
University of Nottingham
School of Biosciences
Sutton Bonington
Leicestershire LE12 5RD   UK

Tel:        +44 (0)115 951 6156
            +44 (0)116 271 4502
Fax:        +44 (0)115 951 6157
email:      arthur.rowe at nottingham.ac.uk
            arthur.rowe at connectfree.co.uk (home)
Web:        www.nottingham.ac.uk/ncmh/business
*******************************************************


From: Marcelo Nollmann <marcnol at chem.gla.ac.uk>
Organization: University of Glasgow
Reply-To: marcnol at chem.gla.ac.uk
Date: Thu, 17 Oct 2002 15:26:13 +0100
Cc: "'rasmb at rasmb-email.bbri.org'" <rasmb at rasmb-email.bbri.org>
Subject: [RASMB] standard for sv


----------------------------------------------------------------------------
------
The older archived RASMB emails can be found at:
http://rasmb-email.bbri.org/rasmb_archives
and current archives at
http://rasmb-email.bbri.org/pipermail/rasmb/
Search All the Archives at:
http://rasmb-email.bbri.org/rasmb_search.html
----------------------------------------------------------------------------
------

Hi everyone,
I am doing sv experiments on a dna-protein complex (complemented with
se), and then trying to identify the species by using hydro to model the
s-value that the possible architectures of the complex would have. I am
mainly analysing the sv data with sedfit (using c(s) analysis). I was
wondering if it was possible to
use a non-reactive standard either in a different cell or mixed with the
sample that would let me have more certainty on the s-values coming out
of the analysis. I thought of obvious things like lysozyme or bsa, or
maybe latex beads. Does anybody have any experience on this? I would
really appreciate any help or reference to previous work. (I am sorry if
the question is too naive)
Thanks a lot in advance,
Marcelo Nollmann
Glasgow University


-- 
"He decidido enfrentar la realidad, asi que apenas se ponga
linda me avisan"
                                             Felipe
("I've decided to face reality. Can anyone let me know as soon
as it gets nicer?")


_______________________________________________
RASMB mailing list
RASMB at rasmb-email.bbri.org
http://rasmb-email.bbri.org/mailman/listinfo/rasmb


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://list.rasmb.org/pipermail/rasmb-rasmb.org/attachments/20021017/94a2e657/attachment-0001.html>