Dear ALL:
Thanks a lot for the valuable guidance.
Here are some updates of my problem.
The current model is 100% complete now. The two good molecules fit into the
density very well.
I am currently using P2221 space group.
I split the data according to suggestions I got using mtzutils.
Imean for K=2n : 885.3 SigImean: 30.1
Imean for K=2n+1: 82.1 Sigmean: 12.2
The weak refelections are about 10% intense of the strong ones.
I ran Phaser with these two data sets and again there is a slight shift (<1A)
between the solutions.
The initial rigid body refinement by Refmac shows the following statistics for
all the four molecules in the ASU.
strong data, k=2n, Rfactor28.05%, Rfree29.7%
weak data, k=2n+1 Rfactor41.47%, Rfree41.04%
If I use all my data, the Rfactor is around 35%.
I tried to refine all the 4 molecules after transforming the two good ones to
get the other two using the vector (0.0 0.5 0.0) revealed by native patterson.
It was still stuck around 35%. It seems that the refinement can not give very
accurate postions for the other two though more or less correct.
How can I proceed? If I reduce the b axis half and reindex the data, I
guess I will get very good refinement statistics. If I return to the larger
unit cell, what will happen? Can the refinement in reduced unit cell give
accurate positions of the other two?
Thanks again for your suggestions.
Jerry
> Date: Mon, 7 Sep 2009 11:22:33 +0100
> From: [email protected]
> Subject: Re: [ccp4bb] Rfactor got stuck with a data having alternate strong
> and weak reflections.
> To: [email protected]
>
> Ben Ammar Youssef wrote:
> > Hi all,
> > I just want to ask another question related to this topic:
> > Based on the example given by Jerry McCully and if the data has
> > alternate strong and weak reflections, how can we split it into two
> > different files; one containing the weak reflections and the other
> > containing the strong reflections?
> >
> > Youssef
> >
> >
> mtzutils
>
> keyword RZONE 0 1 0 2 0 ~ # gives reflections with k = 2n
> RZONE 0 1 0 2 1 # gives reflections with k = 2n+1
> Eleanor
_________________________________________________________________
Hotmail® is up to 70% faster. Now good news travels really fast.
http://windowslive.com/online/hotmail?ocid=PID23391::T:WLMTAGL:ON:WL:en-US:WM_HYGN_faster:082009
