From: Philo, John <jphilo@amgen.com>
  To  : 'rasmb' <rasmb@bbri.harvard.edu>
  Date: Tue, 13 May 1997 08:51:10 -0700

release of program SEDNTERP

RASMBers,

The sedimentation data interpretation program/database SEDNTERP is now
available for downloading from the RASMB software archive.  This program
automates many of calculations needed for reducing raw sedimentation
coefficients, and/or sigma values from equilibrium experiments, into
molecular weights and hydrodynamic parameters (axial ratios etc.),
including estimation of buffer densities and viscosities. This Windows
program is distributed as a self-extracting, self-installing archive,
sdtr9701.exe (1.7 MB).

This program is the result of a collaborative effort between Tom Laue
and his students, David Hayes, and myself.  A more complete description,
taken from 
the descript.txt file found with the program, is attached at the end of
this message.

The program includes an expandable database of coefficients for
calculating the effects of solutes on solvent density and viscosity.
Thanks to Tom's student Ian MacGregor, the database now covers over 90
compounds, including all the data tables from the CRC Handbook of
Chemistry & Physics (except those for various animal urines, which
didn't seem too useful!).  Those data tables and corresponding fits and
graphs are also available as Origin .ORG files combined into a
self-extracting archive "density.exe" found in a "density-viscosity"
subdirectory under the SEDNTERP program directory.  Also included in
that archive are Origin fitting functions and template files so that you
can fit your own density and viscosity data (or re-fit the CRC data over
different concentration ranges) and add those coefficients to the
database in SEDNTERP.

With regard to the density and viscosity data, Tom and I want to
encourage you to make available to everyone any data you have for
compounds other than those already included.  Geoff Howlett, one of our
beta testers, has already started this process by providing data for
trifluoroethanol which are now included in the database.  If you submit
results for other compounds to Tom or myself, we can maintain a text
file with additional coefficients on the RASMB site, and such new data
can be incorporated into future program updates.

We hope you will find this program to be helpful.

'Best to all,

John Philo

--------------------------------
The SEDNTERP sedimentation interpretation/database program is a general 
purpose tool for the interpretation of sedimentation velocity and 
sedimentation equilibrium experiments.  It collects many diverse 
functions into a single program.  The program both uses a database of 
information about amino acids, common salts and buffers, etc., and 
allows the user to build a second database of information about samples,
solvents, experimental results (sigma values from sedimentation 
equilibrium and raw sedimentation coefficients), and interpreted 
results (molecular weights, S20w, f/fo, axial ratios, etc.).  
The program requires Microsoft Windows (3.1 or Win95).

Among the things SEDNTERP will do are:

1) Estimate solvent densities and viscosities for 93 salts and buffer 
components, using interpolation formulas based on data from the CRC 
Handbook of Chemistry & Physics and other sources.  

2) Calculate protein molecular weights, partial specific volumes, and 
hydration based on amino acid composition.  A long list of protein 
conjugates such as sugars, cofactors, and metal ions can also be added 
to the amino acid composition. Partial specific volumes can also be 
corrected for the presence of 8M urea or 6M guanidine hydrochloride.

3) Perform basic manipulations and reduction of raw sedimentation data: 
converting sedimentation coefficients to standard conditions, 
interconversion between buoyant and true molecular weight, and 
converting sigma values from sedimentation equilibrium to molecular 
weight.

4) Derive hydrodynamic parameters from sedimentation coefficients: 
frictional coefficients, minimum frictional coefficients, Stokes 
radius, maximum degree of hydration, and axial ratios for ellipsoid 
and cylinder models.  All such calculations are done using both the 
"vbar" and "Teller" methods.  Diagrams of the derived molecular shape 
and absolute size, and graphs of axial ratio versus hydration may be 
displayed and printed.

5) Sum amino acid sequences to derive the amino acid composition for 
use in vbar or sequence molecular weight calculations.  Sequences can 
be cut and pasted from other sources such as sequence databases, or 
entered manually, and the sequence is stored in the database along 
with other information about that sample.

6) Calculate protein extinction coefficients, charge, and isoelectric 
points based on their composition.  Graphs of charge versus pH 
(titration curves), and of predicted absorption spectra (250-320 nm) 
are also available.

7) Extend the physical constant databases with new information from 
the user.  The user may add data for such things as non-natural amino 
acids, other protein conjugates, and densities or viscosities of other 
solvent components.

8) Store and retrieve information and results about the user's own 
sedimentation experiments.  The user creates "sample" records defining 
a sample name, vbar, MW, and hydration, and named "buffer" records 
defining solvent density, viscosity, and pH.  Then "experiment" 
records are created with experimental sedimentation coefficients 
and/or sigma values for a pair of "sample" and "buffer" records.  
Finally, "result" records can be created for a defined hydrodynamic 
analysis of a given "experiment" record, such as a prolate ellipsoid 
shape analysis using the Teller method.


The SEDNTERP program grew out of the chapter "Computer-Aided 
Interpretation of Analytical Sedimentation Data for Proteins" by Tom 
Laue et al. in the Harding, Rowe, and Horton 1992 book, and it largely 
follows the terminology and computational methods outlined there.  
An extensive Windows Help file is available with the details of the 
methods and computations.

The program code and Help file were originally developed by David Hayes 
in collaboration with Tom Laue.  Over the last two years these were 
then extended and modified by John Philo at Amgen, resulting in the 
current released version.

A small example database with a few defined samples and buffers is 
distributed along with the program to help illustrate its use.