From: Stephen Harding <Steve.Harding@nottingham.ac.uk>
  To  : jphilo@mailway.com
  Date: Tue, 21 Mar 2000 09:21:05 -0000

RE: SEDNTERP update

Thanks John. This might explain why one or two people were 
getting slightly different answers with ELLIPS1 which I think is 
correct (the latter agrees with the limiting values of oblate and 
prolates in the general triaxial programme ELLIPS2 which uses the 
full formulae with elliptic integrals)
Steve
 On 20 Mar 00, at 21:33, John Philo wrote:

> As a follow-up to Tom's message, there is one other bug fix in the new
> version that he didn't mention.
> 
> In prior versions there was sometimes a problem in the graphs of ellipsoid
> axial ratios versus hydration. The graphs for prolate ellipsoids using the
> Teller method or oblate ellipsoids using the vbar method were okay. However
> for the prolate/vbar or oblate/Teller combinations I messed up the logic,
> and hence the data displayed on the graph was actually for the other of
> those two combinations. This is now fixed too. The data shown on the Results
> page were okay.
> 
> Lastly, I wanted to remind SEDNTERP users that if you publish or export the
> graphs showing the shape of ellipsoids (or cylinders) you must be careful
> not to distort the proportion of the X and Y axes. Despite the warning
> messages that are issued about this, figures have appeared in print where
> the X and Y axes are not the same physical length and hence the shape has
> been distorted.
> 
> If you change the X/Y axis length ratio so the graph is no longer square,
> your molecule will appear ellipsoidal even if the actual axial ratio is 1!
> Please be careful when printing those graphs or pasting them into another
> program, following the program guidance in order to preserve the aspect
> ratio.
> 
> John
> 
> -----Original Message-----
> From: Tom Laue [Tom.Laue@unh.edu]">mailto:Tom.Laue@unh.edu]
> Sent: Monday, March 20, 2000 1:21 PM
> To: rasmb@alpha.bbri.org
> Subject: Sednterp update
> 
> 
> Dear RASMBers,
>   While working on the update to Sednterp, John Philo found an error in the
> original code from my lab. In particular, the temperature correction of the
> density for heavy isotopes was wrong: the melting point of the isotope was
> used instead of the temperature of maximum density in the calculations.
>   Some quick testing by John indicates that the miscalculation for pure D2O
> would produce errors of ~.003 in rho. This error, in turn, propogates to a
> ~1.1% error in the mass estimate for a protein with a vbar = 0.73 ml/g.
>   To give you a better idea of the level of imprecision, here are values
> John calculated for pure D20 at different temperatures:
> 
> T	old code	new code	CRC(Kell data, not formula)
> 5	1.10601	1.10564	1.10562
> 10	1.10569	1.10601	1.10599
> 20	1.10392	1.10537	1.10534
> 30	1.10091	1.10325	1.10323
> 40	1.09689	1.09997	1.09996
> 
>   There are two ways to update Sednterp with this problem patched. The new
> full install is the file sdtr0002.exe (1.8 MB). Since only the program
> executable is changed, there is also a file sdnupdat.exe (220 KB) which
> updates versions 1.01-1.04 to 1.05. The update-only file does not have a
> true installation utility; when the self-extracting archive is run you must
> supply the correct folder location for the previous SEDNTERP installation
> (usually c:\sednterp) and the archive will overwrite the older program file.
>   Both the full update and the patches are available from either John
> Philo's site or the RASMB ftp site:
> http://www.jphilo.mailway.com/download.htm
> ftp://alpha.bbri.org/rasmb/spin/ms_dos/sednterp-philo/
>   If you do a full re-install, you must be careful not to overwrite your own
> databases (make a backup of SEDNTERP.MDB and PHYCONST.MDB just in case, and
> answer YES that you want to preserve your existing databases when you get
> the prompts).
>   My apologies to everyone for allowing this error, and my thanks to John
> Philo for finding and fixing it.
> Tom
> --------------------------------------------
> Tom Laue
> Professor and Director of the Center to Advance Molecular Interaction
> Science
> University of New Hampshire
> Biochemistry and Molecular Biology
> Rudman-379
> 46 College Rd.
> Durham, NH 03824-3544
> 
> Phone: 603-862-2459
> FAX:   603-862-4013
> ---------------------------------------------
>