Pavel, I have a problem with the number of reflections for refinement
PHENIX report
a. I have almost complete data for the data set - 48071 reflections. I
keep always anomalous pairs separated.
PHENIX reports (using phenix.model_vs_data.log) 94475 reflections
(reasonable, taking into account 98% data and possible mismatch in
anomalous pairs). I assume that (+) and (-) are separated in this case.
b. Problems come when depositing data to PDB using their tools. They
do not accept ( I think) anomalous separated data and
complain about number of reflections that is not consistent between
reflections file and PDB file header.
c. What happened in the extreme cases? Let us say that for anisotropic
data I have 78% of completeness. The anomalous data is separated.
Will PHENIX use during refinement complete set of reflections ( for
various density maps calculation) or it will use only WHAT IS THERE IN
THE INPUT FILE?
Dr Felix Frolow
Professor of Structural Biology and Biotechnology
Department of Molecular Microbiology
and Biotechnology
Tel Aviv University 69978, Israel
Acta Crystallographica D, co-editor
e-mail: mbfro...@post.tau.ac.il
Tel: ++972 3640 8723
Fax: ++972 3640 9407
Cellular: ++972 547 459 608
On Mar 12, 2009, at 8:27 PM, Pavel Afonine wrote:
Hi Dale,
1) There is a need for additional validation of structure factor
depositions.
PHENIX has tools for this:
1) "phenix.cif_as_mtz" will convert the PDB data file with
diffrcation data into MTZ file. It automatically will figure out if
the data are X-ray: Iobs or Fobs, or Neutron Fobs or Iobs.
2) The next step will be running "phenix.model_vs_data"
that will take the MTZ from from above and the corresponding PDB
file and give a complete statistics, that you can immediately
compare with the published value.
Note, phenix.model_vs_data can handle:
- twinned data;
- neutron data;
- all unknown ligands dictionaries are generated internally;
- PDB files with multiple models (with multiple MODEL records).
I run it every month or two, and so I have a nice list of
"interesting cases". The database of all converted to MTZ data files
is used internally by PHENIX developers for various developments
etc...
In fact, this was used in POLYGON validation tool: Acta Cryst.
(2009). D65, 297-300
Pavel.
