[phenixbb] ncs

Dale Tronrud det102 at uoxray.uoregon.edu
Tue May 5 14:44:35 PDT 2009


   If you have nearly C3 ncs symmetry parallel to the crystallographic
3-fold, you very likely also have some form of pseudo-translational
symmetry.  If present this will lead to quite a few of your intensities
being very small, and poorly measured.  This situation can require you
to continue restraining ncs even at 2.1A resolution, particularly if
many of your weak reflections are "unobserved".

   It is quite possible that the free R is telling you exactly what
you need to know, that your structure factors alone are insufficient
to define your model.

   Dr. Echols' comment:

>> It may be possible that the NCS restraints are biasing your test set,
>> leading to the suspiciously low Rfree-Rwork gap, but I thought this
>> only applied when the NCS approached crystallographic symmetry, and

is the reverse of my understanding.  The cross-talk between the working
set and test set is minimal when the ncs approaches crystallographic
symmetry, and at its worst when the two are unrelated.  When the ncs
and "cs" are similar the ncs images of a test set reflections falls
very near its proper symmetry images, which are also in the test set by
definition.  When the two types of symmetry are unrelated, the ncs image
of a test set reflection will fall near a reflection which only has a
5% chance of being in the test set.  Now you have a direct relationship
between a reflection in the test set and one in the working set.

   With care, a test set can be selected that minimizes this crosstalk.
Since you have ncs symmetry that is masked by cs symmetry this is not
much of a problem for you.  (I don't quite understand how the 2 copy
ncs of the DNA relates to the crystallographic symmetry, so it might
be a problem.)

Dale Tronrud



Maia Cherney wrote:
> Thank you Nat, Pavel,
> 
> Actually, the NCS approaches a 3-fold crystallographic symmetry axis 
> with 3 types of protein molecules plus 2 restraint groups for DNA molecules.
> 
> The result is the same with "optimize_wxc=true optimize_wxu=true" 
> options, Rfree is going up without NCS by ~1.5%. Rwork is practically 
> the same.
> 
> Should I increase the test set (now it's 5%)? or should I leave NCS=true 
> for the final pdb?
> 
> Maia
> 
> 
> Nathaniel Echols wrote:
>> On May 4, 2009, at 6:48 PM, Maia Cherney wrote:
>>   
>>> The resolution is 2.15 A. The NCS was always on during the refinement
>>> until we got low R factors (19.2% and 21.2%). Then the NCS was turned
>>> off for the final refinement and the R factors increased, which is
>>> strange as they should be going down when you apply less restraints.
>>>     
>> This isn't necessarily true - regardless of which program you use to  
>> refine, R-work isn't actually the refinement target, and minimization  
>> algorithms aren't foolproof either.  When you remove the restraints  
>> you are effectively decreasing the observation:parameter ratio, which  
>> increases the risk of overfitting.  If nothing else I would expect the  
>> gap between Rwork and Rfree to increase, but I've also seen both  
>> increase when NCS restraints were removed.
>>
>> It may be possible that the NCS restraints are biasing your test set,  
>> leading to the suspiciously low Rfree-Rwork gap, but I thought this  
>> only applied when the NCS approached crystallographic symmetry, and  
>> there's no such thing as fivefold crystallographic symmetry.   
>> Hopefully someone else on the list can clarify this.
>>
>> -Nat
>>
>> -------------------
>> Nathaniel Echols
>> Lawrence Berkeley Lab
>> 510-486-5136
>> NEchols at lbl.gov
>>
>>
>>
>> _______________________________________________
>> phenixbb mailing list
>> phenixbb at phenix-online.org
>> http://www.phenix-online.org/mailman/listinfo/phenixbb
>>
>>   
> 
> _______________________________________________
> phenixbb mailing list
> phenixbb at phenix-online.org
> http://www.phenix-online.org/mailman/listinfo/phenixbb



More information about the phenixbb mailing list