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


> Something I have never found to work to my satisfaction has been solvent
flattening after molecular replacement where the model is somewhat
partial (i.e. more than 20% of the new structure is not modeled by the
MR solution).  In a case I was trying out for someone else a few months
back I used DM with a variety of automatic and more manually-generated
solvent masks, and this produced no significant gain in the phases to my
eyes.  So I am left with the suspicion that I'm not using the best
methods.

You don't mention which resolution you're working with, which probably
makes a big difference.  I tryed this out with a ~4 Angstrom dataset,
using a mask that worked well for the experimental phases; solvent
flattening on model phases didn't show any sign of density not present in
the model (given the resolution, I wasn't expecting it to; but you're
probably using higher-resolution data).

> Does anyone have any advice ? - I've found Coulombe and Cygler's 1997
paper (Acta Cryst D53, 426-433) - but I imagine that there is other
information out there that an hour's Googling has yet to turn up.

Other than not trying it <g>...give pirate a shot with a couple of
different reference sets, or use the partial model to make it easier to
find HA sites for experimental phases.

Pete

Pete Meyer
Fu Lab
BMCB grad student
Cornell University




Reply via email to