[RASMB] s20w calculation

[John Philo]

David, I appreciate the nice things you said about SEDNTERP 1, but people
need to remember that it took quite a long time to get the bugs out of that
version (I think a decade or more!). The calculations are the easy part, it
is the user interface that creates the problems, and that is 100% new in
version 2. So the community needs to have some patience about version 2.
 
Remember too that SEDNTERP 1 got to where it was only because the community
took the time to report and properly document the bugs so they could be
tracked down and corrected. It also only to got to where it was due to the
major contributions of David Hayes and Tom Laue.
 
Best to all,
 
John
  _____  

From: rasmb-bounces at list.rasmb.org [mailto:rasmb-bounces at list.rasmb.org] On
Behalf Of David Hayes
Sent: Friday, August 09, 2013 2:05 PM
To: rasmb at rasmb.org
Subject: Re: [RASMB] s20w calculation


Hi John,


I sent some off list response with extra information and opinion I am not
sure about and don't want to be on list.
I am also waiting for Walter or Peter Sherwood to answer about SEDANAL
questions, but I suspect that you are thinking SEDANAL is doing s20w
corrections that it really is not, but I don't know if that is the problem.


Everyone should know that the Web SEDNTERP and the desktop beta share the
exact same calculation source code, so if they are the same version, they
better agree on everything, feature for feature, bug for bug.


But also, basically trust the by hand calculations as was posted previously,
ask the programmers which numbers are s20w and which are not, and thank you
for posting the problems so the developers of SEDNTERP can work out the
bugs.  Due to John Philo's incredible and meticulous work, SEDNTERP 1.2 was
incredibly reliable and calculated things to 5 decimal places reliably and
exactly the same as the published equations.  the new version of SEDNTERP
will take a lot of sweat to get to that level of excellence.


David Hayes

  _____  

From: John Sumida <jpsumida at u.washington.edu>
To: drdavidbhayes at yahoo.com 
Cc: rasmb at rasmb.org 
Sent: Friday, August 9, 2013 1:44 PM
Subject: Re: [RASMB] s20w calculation


Hi David,
 
Thanks for taking the time to respond. I'm attempting to respond to your
post. Apologies if I'm committing some type of etiquette violation in
responding to the group, but as you requested some clarification as to my
process I thought to respond publicly.
 
To answer your questions.
 
I've been trying to understand why this particular polymer material produces
a large s20w value when another material with different surface charge
properties produces an s20w of 7.9s and sediments very differently.
 
So the back calculation, going from s20w to s*, was me playing with the data
to see if I could go from the 17.3 s20w back to s* which I would have hoped
- as you point out - would not have been much different from 17.3s.
 
My initial calculation using the s20w correction equation  (equation 19) was
comforting in that the calculated s* was 16.4s.  I used SEDFIT to perform
the same calculation just to check that all was in order - and as I stated
SEDFIT returned the 16.4s value as had the calculation and the analysis in
SEDANAL.  Note that vbar and solvent parameters had been entered in into the
experiment file and gOfs files had been generated by running DCDT.  So I
believe the s values returned in SEDNAL are s20w values.
 
The problem arose when I used the Experiment page in SEDNTERP (version
20120828 BETA). Using the calculated value of s*, SEDNTERP did not return
the same value for s20w that SEDANAL output. 
 
I also attempted the same operation using the web version of SEDNTERP which
resulted in the same result, namely calculating an s20w of 27.8s when the
experimentally observed s20w was17.3s.
 
I had not thought to turn off the s20w calculation (in order to do so do I
just remove vbar from the experiment info?) and attempt to fit for s*.  Will
try that next.
 
Best regards,
 
John Sumida, Ph.D.
 <http://depts.washington.edu/cidb4bio/index.shtml> Analytical Biopharmacy
Core Facility
University of Washington
 <http://www.moles.washington.edu/> Molecular Engineering & Sciences
Institute, G22
3946 West Stevens Way NE
Seattle WA 98195-1653


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://list.rasmb.org/pipermail/rasmb-rasmb.org/attachments/20130809/ebc24c67/attachment.htm>