solvent flattening

[Jay Pandit]

John Kuszewski (johnk at loglady.ninds.nih.gov) wrote:
: Could anyone give me a quick explanation of what 
: solvent flattening is?

: -- 
: John Kuszewski
: johnk at spasm.niddk.nih.gov
: I'm not an idiot, but I play one on USENET.

 Here is areally quick explanation:
 Solvent flattenning is a way to "improve" experimentally calculated
 phases. You start with a set of reflection data and calculated phases
 from which you compute an electron density map. If the phases are poor,
 your map will be noisy, and it might be difficult to follow the protein
 backbone. However, if you can make out at least the outline of your
 molecule ( ie the protein/solvent boundary ) then this information can be
 used to improve the phases. All the density outside the boundary is set to
 zero, or some low constant, and the new density distribution is
 back-transformed to give a new phase set. This is then "combined" with
 the starting set, to
 give an improved phase set. The process is repeated over many iterations.
 FOr details, read B.C. Wang's article in Methods in Enzymology v115.

 Cheers,
Jay
--
-----------------------------------------------------------------------------
 Jay Pandit
 Central Research Division
 Pfizer Inc                                     Phone: 203 441 3738
 Eastern Point Road                             Fax  : 203 441 4111
 Groton CT 06340                                email: pandit at pfizer.com
-----------------------------------------------------------------------------