You know you can convert the CNS FreeRflags to the mtz file?
Use Reflection utilities / convert to mtz and say you are inputted CNS /
Xplor data
Eleanor
But I believe there is an inevitable difference between CNS and Refmac
FreeRs - CNS scaled the work and free data seperately which for lower
resolution data will give a reduced difference between the 2 values.
Eleanor
Zheng Zhou wrote:
Thank you all for fast replying. The reason that I am trying to use a
different set of FreeR flag from the very beginning of the refinement is
that for some data set, my colleague's CNS refinement gave converged Rfree
and R work, 3% difference. However both my CNS and CCP4 refinement gave a
difference of about 7%. I was trying to use the same parameter settings as
my colleague in simulated annealing. His program is from his previous lab
and all the input files for CNS are modified as a batch file. But mine are
just individual files downloaded from CNS website.
Among those things I tried, I thought about looking at the true input data
for computing Rfree and Rwork. It appeared to me the FreeR flags in my
"First" runs are never randomized. Thanks for your reply, I guess I could
use a different flag number other than 0 (I know I am not supposed to change
this in the middle of a refinement). I even tried to use my colleague's
cross-validation file .cv for CNS. It still diverge in my runs. Anyone met
similar problem before, where CNS and CCP4 give a quite different R factors?
Thanks for your insight.
Zheng (Joe)
On Jan 23, 2008 9:50 AM, Zheng Zhou <[EMAIL PROTECTED]> wrote:
Hi, All
Could any tell me how CCP4 handle free R flag? I know It is important to
select the same** FreeR reflections if I move to next step of refinement.
But everytime I start from fresh, the freeR Flag remains unchanged. The
Rwork and Rfree of my models are fine ( 20.7% and 22.9%). I thought Rfree
was randomly generated. I asked more experienced people in the lab and
neighbor labs, and didn't get a straight answer. Did I do something wrong?
Thanks and sorry to bother others.
Zheng (Joe)
