[phenixbb] test set count question
fyabukar at embl.fr
Wed Aug 8 14:43:26 PDT 2012
Thanks very much for the answers!!!
Indeed I missed to include in my message that xdsconv outputs not only
h,k,l,F,SigF,DF,SigDF, but also the isym,i - Im sorry.
isym=0: both F(+) and F(-) were observed.
F=(F(+) + F(-))/2; DF=F(+) - F(-)
isym=1: only F(+) was observed.
F=F(+); DF and SigDF are omitted from the
isym=2: only F(-) was observed.
F=F(-); DF and SigDF are omitted from the
So it seems to be OK. Since I refine directly SAD data i find
convenient to have also the DF/SigDF in order to make anomalous
difference maps if needed.
Its just ive never seen a pdb header with this "doubled" test set count
column. Maybe most of the people use only averaged data (<F> <SigF>) for
refinement whatever the data (SAD or Native).
Thanks a lot once again!
On 2012-08-08 20:34, Pavel Afonine wrote:
> I've seen files with bare F SIGF DANO SIGDANO. Not sure what Filip
> Yes, if instructions about how to reconstitute +/- reflections are
> provided then that all right.
> On 8/8/12 11:26 AM, Nathaniel Echols wrote:
>> On Wed, Aug 8, 2012 at 11:19 AM, Pavel Afonine <pafonine at lbl.gov>
>>> <Fobs> or "F/SigF and DF/SigDF" are derived from Fobs(+) and
>>> Fobs(-) with
>>> some information lost. That is you cannot restore the original
>>> Fobs(+) and
>>> Fobs(-) from <Fobs> or from "F/SigF and DF/SigDF". phenix.refine
>>> Fobs(+) and Fobs(-) from "F/SigF and DF/SigDF" for internal use
>>> making an
>>> arbitrary decision about the sign. phenix.refine cannot handle
>>> "F/SigF and
>>> DF/SigDF" and has no reason to do this.
>> Does XDS (or xdsconv) not include the ISYM column? In some files,
>> get a combination of columns like this:
>> F SIGF DANO SIGDANO ISYM
>> which is the same thing, but the ISYM column indicates how the DANO
>> applied to the Friedel mates, meaning that we can reconstruct the
>> original F+ and F-. There's nothing wrong with using these data in
>> Phenix - without ISYM, however, the DANO is useless.
>> phenixbb mailing list
>> phenixbb at phenix-online.org
> phenixbb mailing list
> phenixbb at phenix-online.org
More information about the phenixbb