Hi Tassos,
Am 11.12.2008 um 13:58 schrieb Anastassis Perrakis:
...
Having said these, I will shamelessly admit here (as I think I also
did in the latest Gordon conference?) that I have never found an
omit map to be *really* useful,
and it never told me something I could not see in the 2fo-fc and fo-
fc maps. Have people seen recently (post-likelihood) an omit map
showing something different than the combination of the two 'usual'
maps? If so I would be really interested to see them.
You can send me the mtz and pdb's, and I promise to put the pictures
on a web site together with an apology for writing this paragraph.
I leave out questionable parts until every other part of the structure
including waters and ligands are satisfactorily well refined. In my
experience, the resulting maps for the questionable parts look usually
better interpretable than having put some model tentatively into the
density with all its resulting local model-bias. I had one case at 3.2
A resolution with a somehow strange looking important active site loop
that suddenly became clear to me, with a more reasonable
interpretation than before, after this omit-procedure mentioned above.
However, when I looked back to the former maps, maybe, I could have
found this interpretation as well without an omit-map, although the
omit-maps "looked" clearer to me. I think, that often subtle
differences can suddenly make an otherwise difficult interpretation
clear to the brain ...
Unfortunately, I don't have the different stages of refinement and
maps of that example at hand anymore.
Best regards,
Dirk.
*******************************************************
Dirk Kostrewa
Gene Center, A 5.07
Ludwig-Maximilians-University
Feodor-Lynen-Str. 25
81377 Munich
Germany
Phone: +49-89-2180-76845
Fax: +49-89-2180-76999
E-mail: [EMAIL PROTECTED]
*******************************************************
