Discussion List Archives

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]


Yesterday's posting to sa_scat from Bob Lagasse looking for a reference 
manual to a software package named CONTIN prompted me to power up trusty 
old Sherlock to go looking for information out on the WWW. I found a site 
with the URL 


Whose first few lines read as follows:



*	Inversion of data represented by linear algebraic or integral 
equations, such as general analysis of multi-exponential decay (Laplace 
inversion) or estimating secondary structure from circular dichroism; 
*	Most aspects of the problem can be easily modified in small 
user-oriented modules; too many applications to list here have been 
implemented by users; 
*	Running in hundreds of laboratories with a wide variety of computers. 


While this site did not say anything specific about the manual that Bob 
was seeking, it does appear to be a contact that should help him. I also 
got some mail on this from other people. This is the disadvantage of 
having sa_scat as a moderated listserver: I get a certain amount of 
e-mail because people are not free to post directly to the list.

Being (relatively) new to the SAS business, I was not aware of the CONTIN 
code until last evening. However, a quick perusal of its aplicability 
indicates to me that members of the SAS community may have much better 
choices. The varied works of Dmitri Svergun  (see 
http://www.embl-hamburg.de/ExternalInfo/Research/Sax/), Otto Glatter's 
IFT and GIFT procedures (see 
http://physchem.kfunigraz.ac.at/sm/Personal/Otto/Otto.htm), and the 
review in the literature by Jan Skov Pedersen (Advances in Colloid and 
Interface Sciences, V70 (1997) pp 171-210) suggest that there are ways to 
analyze SAS data that may be much better suited to our needs. It seems to 
me that data inversion procedures should incorporate as much specific 
information as they can about the experiment.

Both Dmitri and Otto gave very nice overviews of their work at last 
week's ACA Small-Angle SIG meeting. I hope that their work will soon 
receive wider recognition in the New World.

I would like to encourage further discussion on this matter. The issue of 
efficient extraction of information from SAS data and assessing the 
precision of the resulting estimates is one that needs a lot more study. 
Please feel free to post comments to the listserver.


Dr. John D. Barnes               email: john.barnes@nist.gov
Natl Inst of Stds and Tech       Voice: 301-975-6786
100 Bureau Drive, Stop 8544        FAX: 301-975-4977
Gaithersburg, MD 20899             URL: http://www.nist.gov/sas