[phenixbb] phenix.refine and R free values

Nathaniel Echols nechols at lbl.gov
Wed Sep 3 08:41:57 PDT 2014


On Wed, Sep 3, 2014 at 8:33 AM, PC <patrick.cossins at inbox.com> wrote:

>  I see in 2y78.mtz (example I used below) has both in the MTZ:
>
>   F(+),SIGF(+),F(-),SIGF(-)
>   I(+),SIGI(+),I(-),SIGI(-)
>
> I am confused, why? One is sufficient isn't it?
>

"Why" is "because that's what the crystallographer submitted to the PDB".
 It's actually better to have both arrays: the intensities are (usually)
the original data as measured on the detector, while the amplitudes have
been processed such that negative intensities are either removed (in older
versions of Phenix) or adjusted based on the French-Wilson protocol (in
newer versions, and also in CCP4 utilities), and therefore do not exactly
capture the original data.  But the amplitudes are what the model was
actually refined against, so to have a complete record of the experiment,
it's helpful to have them too.  (You're correct, however, that to just
re-run phenix.refine we only need one of these.  Note that since the
original structure was refined with an older version, they may have used
fewer reflections because the negative intensities were discarded.)

-Nat
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://phenix-online.org/pipermail/phenixbb/attachments/20140903/1951a06e/attachment.htm>


More information about the phenixbb mailing list