[RASMB] SedAnal: Sedimentation Velocity Analysis package called SedAnal

[Walter Stafford]

Dear RASMBers;

	A new software package called SedAnal has been developed at 
BBRI and has been made available on the RASMB FTP archive site.

SedAnal will perform curve fits to sedimentation velocity data 
obtained from interacting (and non-interacting) systems. It can fit 
for sedimentation coefficients, molecular weight. loading 
concentrations and equilibrium constants. It is capable of  fitting 
up to 3 thermodynamic components, and up to 7 species  related by up 
to 6 chemical reactions. It can fit globally data from up to 5 cells, 
simultaneously.  It runs under Windows98 and has an easy to use 
graphical user interface. It is accompanied by a User's Manual.

It fits to time difference data to eliminate time-independent, 
radially-dependent systematic noise completely. It has a preprocessor 
section that allows choice of the meniscus and base positions as well 
as the radial range of data to use for fitting. The preprocessor 
allows essentially complete removal of radially-independent, 
time-dependent optical "jitter" and correction for  integral fringe 
shifts associated with interference data.

The package contains a Model Editor that allows the user to develop a 
database of reaction schemes which can be used to fit the data.

The fitting section ( Control Screen) allows the user to choose one 
of the models, select and enter initial values for the parameters to 
be fit (or held) and select the cells to be combined for global 
fitting.

The model editor allows one to create various reaction schemes 
ranging from a single-component, single-species system (i.e. no 
interactions) to a single-component, 2 species system (e.g.. simple 
monomer, n-mer self-association, 2A=A2), to a 3 component, 3 species 
system (i.e. simple complex formation of the type A + B = C), a 2 
component, 4 species system (i.e. simple complex formation with self 
association of one of the reactants or the complex (e.g. A+B=C; 
2A=A2), a 2 component, 4 species system like an antigen-antibody 
reaction (e.g. A+B=C;C+B=D) and so on.

The current version of SedAnal can handle up to 3 components, 7 
species and 6 chemical reactions  in any combination consistent with 
the constraint that Number_of_components = number_of_species - 
number_of_chemical_reactions, of course. (This is probably more than 
one should ever dare to use without some prior knowledge). Isodesmic 
self-association will be handled in a future version.

The Control Screen also allows the user to generate synthetic data 
using any of the models.

The internal routine uses numerical solutions to the Lamm equation 
solved using Claverie's finite-element method. Mass action is 
accounted for by various kinetic schemes and is solved internally by 
the Bulirsch-Stoer differential equation solver routine. Least 
squares minimization is done by the Simplex method of Nelder and Mead.

There is an user's manual in PDF format. The manual is fairly 
extensive and still evolving.

The software/documentation package can be downloaded from the RASMB 
FTP archives at

ftp://rasmb.bbri.org/rasmb/spin/ms_dos/sedanal-stafford/

The zip archive contains the user manual, the model editor, an 
example Model database file, Sedanal.exe, which is the interface to 
the Preprocessor and the Fitter Control Screen. Please read the 
README file and the user's manual first. AND take note of the 
mandatory directory structure. The unzipping process should put 
everything into the correct directory tree structure.

There is a separate email discussion group at 
http://www.rasmb.bbri.org/mailman/listinfo/sedanal
for those who want to discuss the more arcane aspects of using 
SedAnal without cluttering up the RASMB email list.






-- 
-------------------------------------------------------------------------------------------------------
Walter Stafford
mailto:stafford at bbri.org
direct dial:    617-658-7808
receptionist: 617-926-8040
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://list.rasmb.org/pipermail/rasmb-rasmb.org/attachments/20021017/88263d6d/attachment.htm>