[ccp4bb]: Conclusion of bulk solvent corrected fo-fc map

[Mark DePristo <mdepristo@cryst.bioc.cam.ac.uk>]

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

Hi everyone,

I've done some more work with the fo-fc maps from refmac and xfit.  One 
definite conclusion (thanks to Ian Tickle) is that the default B-factor 
restraints in refmac are definitely too tight (and incorrect in the 
HTML documentation).  Loosening them up not only improved the map 
quality but also dropped down the R and R free by 0.5%.  I've tried a 
newer version of refmac (v. 5.1.29) but the .24 and .29 maps are 
virtually identical.

As several people have suggested, the refmac fo-fc map contains more 
information than the xfit fo-fc map, and is therefore 'noiser', which 
originally lead me to conclude that the refmac fo-fc map was worse.  A 
large contribution to the noise in the refmac fo-fc map are peaks 
floating in space, away from any atoms, including a large number off in 
the solvent.  This noisiness of the refmac fo-fc map, although perhaps 
not a problem for humans, has been a major problem for the automated 
structure determination software we're developing.  Our program would 
much prefer the starker contrasts of the xfit maps than the more 
informative but also more complex refmac maps.  Perhaps a simple 
masking or smoothing procedure could clear up the refmac maps.

Thanks for all of the help,

Mark

Mark DePristo
Ph.D. Candidate
Dept. of Biochemistry
Cambridge University
mdepristo@cryst.bioc.cam.ac.uk
http://www-cryst.bioc.cam.ac.uk/~mdepristo/