From: jack correia <jcorreia@fiona.umsmed.edu>
  To  : rasmb@bbri.harvard.edu
  Date: Mon, 01 Jun 1998 11:41:04 -0500

Re: dc_dt README file correction

rasmber's,

	Walter's s/D method was originally described in an abstract where the
equation reported is correct (Stafford, 1996 Biophys J 70, A231).  It also
appears in more detail in an article Preston Hensley & Walter have on
antibody-antigen interactions, available at a web I believe (Walter?).  To
my knowledge (others?) it has been used only once otherwise, on a His
tagged human kinesin construct K332 (Rosenfeld et al (1996)JBC 271, 30212),
where is was compared with Svedberg.  Both methods gave identical results,
about 3% smaller than the actual MW.  This construct irreversibly
aggregated during sed equilibrium experiments and thus Svedberg or the s/D
methods were essential.  

	This methods fails if any aggregation or asociation occurs where there is
a mixture of species and thus a larger apparent Diffusion coefficient.
Walter say you must be 2 orders of magnitude above K for this to work with
reversible association. Multiple gaussian fits may work if the species are
noninteracting and well separable such that good estimates of D can be made. 


--------------------------------------------------------
 Dr. John J. "Jack" Correia			 
 Department of Biochemistry                    
 University of Mississippi Medical Center       
 2500 North State St.                           
 Jackson, MS 39126                              
 (601) 984-1522                                 
 fax (601) 984-1501                             
 email: jcorreia@fiona.umsmed.edu               
 homepage: http://fiona.umsmed.edu/~biochem/correia.html  
---------------------------------------------------------