*** For details on how to be removed from this list visit the ***
*** CCP4 home page http://www.ccp4.ac.uk ***
I believe you should keep NCS restraints, but release them selectively
where there are clear violations. 6-fold ncs gives you a tremendous
increase in the effective data/parameters ratio, which can be critical
for successful refinement at low resolution.
Using NCS restraints will definitely lower R-free. I believe this is mainly
because of the above-mentioned improvement in effective data/parameters,
refinement goes better, overfitting is restrained, and R-free does what it
is supposed to and tells you you have a better structure. However the
possibility has been raised that enforcing NCS biases the R-free because
reflections in the "free" set are NCS-related to reflections in the
working set.
You can identify NCS violations by refining without NCS and seeing what
changes (superimpose with/without models and locate outliers) or superimpose
the different chains and locate outiers. Or examine all residues where
procheck reports poor correlation with density, rebuilding individually
in each chain and releasing NCS.
Ed
Vineet Gaur wrote:
Hi all
i m trying to solve the structure of a protein crystallized in p41 space
group by molecular replacement. it is a hexameric protein. each
asymmetric unit have two trimers. i m solving the structure at 3.2A, by
applying NCS between the six protomers. present status of refinement is
Rfree: 30.60% and Rcryst: 30.30%, with a fairly good steriochemistry.
Among the noncrystallographically related molecules i can see no
difference in the main chain but considerable differences in the
sidechains. At this stage of refinement with strict NCS i m not able to
refine the structure further. i m not able to judge weather i should
remove the strict NCS and start refining the individual subunit. so, how
should one decide when to remove the NCS and start refining the
individual subunits.
My second problem related to the NCS is, whenever i remove NCS n start
minimizing the pdb obtained after removing NCS, the Rfree start shooting
up very rapidly. is it bcoz of larger differences among the NCS related
molecule or bcoz of poorly refined structure. pls suggest me what should
i do. I am using CNS for refinement.
Thanx in advance
Vineet Gaur
