Re: [ccp4bb]: amore

[Manfred Buehner <buehner@biozentrum.uni-wuerzburg.de>]

***  For details on how to be removed from this list visit the  ***
***          CCP4 home page http://www.ccp4.ac.uk         ***

On Tue, 11 Mar 2003, Frederico Moraes Ferreira wrote:

> ***  For details on how to be removed from this list visit the  ***
> ***          CCP4 home page http://www.ccp4.ac.uk         ***
> 
> Dear All,
> Let me ask you just a little question about Amore. I have a homodimer
> in the ASU. I tried to look for one monomer, fix it and then tried for 
> the second.
> The scores for the dimer search directly are higher and presents some 
> contrast. My question is:

in most cases, the larger the search model the better the
results. Because there is more structure (intramolecular
vectors) to compare. The larger the "radius" of your
search model, the better the ratio of (productive)
intramolecular to (noise) intermolecular vectors and thus
the signal-to-noise ratio.

However, since there is always a possibility that the
agreement of model and structure is poor (e.g. due to
differences in dimer formation), it is recommended to
check the monomers as well.

Of course, Eleanor is right with her caveats about
special cases such as coincidence with crystallographic
axes.


> Will the highest scores always give the best solution?

it is my (everybody's) first assumption, but remember:
Progress in science works through the unexpected ...
Always check all solutions that appear plausible.


Manfred
----------------------------------------------------------
 Dr. Manfred Buehner             Phone:  +49-931-888-4100
 Physiologische Chemie           FAX:    +49-931-888-4150
 Theodor-Boveri-Institut fuer Biowissenschaften
 Biozentrum der Universitaet Wuerzburg
 Am Hubland,  D-97074 WUERZBURG,  Germany
 Email:  buehner@biozentrum.uni-wuerzburg.de
----------------------------------------------------------