[phenixbb] full omit maps

Miguel Ortiz-Lombardia molatwork at yahoo.es
Fri Dec 29 12:49:17 PST 2006

Hash: SHA1

Hi Tom,

Thank you very much for checking this and for the explanation. If I
understand correctly a full omit map like the one I wanted (with no
rebuilding) might be useful to check, possibly at the end of refinement,
whether there is significant bias. This should give similar information
than the sfcheck omit map.

However, we wanted to check a particular region of the model where the
density is poor and not really improving with refinement. Hence, we
wanted to ponder if there was bias but with the aim of building a new
model for that region if required. Am I correct at deducing from your
explanation that for this case it would be preferable calculating the
full omit *with* iterative model-building?



En/na Thomas C. Terwilliger ha escrit:
> Hi Jianghai,
> The phenix full_omit map, if calculated with no model-building, is similar
> to both the SFCHECK map and the CNS composite SA-OMIT map.
> The phenix full_omit map removes atoms in each omit box and in a 2 A
> radius around the omit box, then calculates phases, then puts all the
> omitted regions together.  This is like the CNS SA-omit map with no
> simulated annealing.  It is also like the SFCHECK OMIT map, except that
> SFCHECK does not remove atoms 2 A around the box.
> The real power of the full_omit map is to calculate it WITH iterative
> model-building. In this case the atoms that get built inside the omit
> region get an occupancy of 0.0 each cycle.  The model outside the omit
> region gets better and better, so that the omit density gets improved
> every cycle.  Then when all the omit regions are recombined, the map is
> totally unbiased, but it can be much better than the starting omit map.
> -Tom T
>> Is this map a total OMIT map like SFCHECK does or a SA-OMIT map like
>> CNS does?
>> Jianghai
> _______________________________________________
> phenixbb mailing list
> phenixbb at phenix-online.org
> http://www.phenix-online.org/mailman/listinfo/phenixbb

- --
Miguel Ortiz Lombardía
Centro de Investigaciones Oncológicas
C/ Melchor Fernández Almagro, 3
28029 Madrid, Spain
Tel. +34 912 246 900
Fax. +34 912 246 976
e-mail: molatwork at yahoo.es
- ----------------------------------------------------------------------
Et ainsi ne pouvant faire que ce qui est juste fût fort, on a fait que
ce qui est fort fût juste.
                                                Blaise Pascal, Pensées
Version: GnuPG v1.4.5 (Darwin)


More information about the phenixbb mailing list