Ha, I've been busted by the Phenix team !
To tell the truth I still want to explore the best building and
refinement options. ArpWarp vs Textal and Refmac versus phenix.refine
and just see what gives the best result. I have used phenix.refine
already....it is very good stuff indeed.
Many thanks
Petra
Pavel Afonine wrote:
Hi,
Just to explaine: it is "assumed" that if you came from using phenix
AutoSol and AutoBuild then for refinement you use phenix.refine
(refinement program in PHENIX). phenix.refine uses 1 for TEST and 0
for WORK (similar to CNS). It is not clear to me at all why you need
to do the swap.
Pavel.
Petra Lukacik wrote:
I have a mtz file (output from phenix AutoSol and AutoBuild) where
the FreeR flag for the test set has a value of 1 and and the working
set has value 0. This is opposite to the ccp4 default where the FreeR
set used within refinement is flagged as 0. Is there a way to swap
the two around so that my file has the ccp4 default arrangement?
Preferably I would like to avoid conversion to ASCII reflection file
formats (and back to mtz).
Many thanks
Petra
------------------------------
Dr Petra Lukacik
NIDDK, NIH
Building 50, Room 4507
50 South Drive
Bethesda
MD 20892
USA
Tel: 301 594 9231
-----------------------------
