[RASMB] time to reach equilibrium

[Borries Demeler]

Dear Patrick,

UltraScan has a module that may be helpful for you (see attached
screenshots).  You can predict the sedimentation and diffusion
coefficients for a number of hypothetical shapes (spheres, prolate
and oblate ellipsoids, and long rods, see attachment 2), as well as
for more complicated shapes using J. Garcia de la Torre's software
HydroPro. Given the hydrodynamic parameters, you can then simulate the
finite element solution for the equilibrium run and find out when the
change from one iteration to the next is less than a certain criterion
(for example, the total change in OD is less than 10e-5 within one hour
or so. That would certainly be in the level of experimental noise of the
XLA. Given that you predicted the s and D for your molecule correctly,
you can then predict for a given column height and position the time it
takes to reach equilibrium.  The UltraScan software allows you to pick
a number of speeds or sigmas and generate the equilibrium profile for
the entire equilibrium run (see attachment #2).

We routinely use it to predict the equilibrium times for all speeds
and simply program them into the XLA. To make sure we actually reached
equilibrium, we take another scan an hour later and compare. We also
generally overpredict the f/f0 value by 10-20% for the molecule under
investigation to be on the safe side, and it gives excellent results. One
word of caution for the rare system that has very slow self-association
kinetics, which are even slower than establishing diffusion/sedimentation
equilibrium, but I haven't seen one like this in a long time.

Regards, -Borries

> 
> Dear RASMB mailing list members,
> 
> A general question regarding sedimentation/diffusion equilibria...
> What is, in your experience, the most effective way to determine and/or
> predict the time needed for the equilibrium to be reached at a given rotor
> speed?
> In practice, do you consider that the equilibrium is attained when no
> change can be observed in the concentration boundary for a certain length
> of time, or when the equilibrium is theoretically reached with a certain
> confidence interval?
> 
> Many thanks for your replies!
> 
> Sincerely yours,
> Patrick England
> 
> ---------------------------------------------------------------------------
> Patrick ENGLAND
> 
> Responsable de la Plate-forme de Biophysique des Macromolécules et de leurs
> Interactions,
> Département de Biologie Structurale et Chimie,
> Institut Pasteur,
> 28, rue du Docteur Roux,
> 75724 Paris Cedex 15, France
> 
> tel: (33-1) 45.68.83.83
> fax: (33-1) 40.61.30.43
> e-mail: biophysique at pasteur.fr
> 
> 
> _______________________________________________
> RASMB mailing list
> RASMB at rasmb-email.bbri.org
> http://rasmb-email.bbri.org/mailman/listinfo/rasmb
> 


(attachment; filename="equiltimes1.gif")
(attachment; filename="equiltimes2.gif")

-------------- next part --------------
A non-text attachment was scrubbed...
Name: equiltimes2.gif
Type: image/gif
Size: 27563 bytes
Desc: GIF image data, version 87a, 776 x 495
URL: <http://list.rasmb.org/pipermail/rasmb-rasmb.org/attachments/20030604/56e8af93/attachment.gif>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: equiltimes1.gif
Type: image/gif
Size: 43469 bytes
Desc: GIF image data, version 87a, 808 x 794
URL: <http://list.rasmb.org/pipermail/rasmb-rasmb.org/attachments/20030604/56e8af93/attachment-0001.gif>