[phenixbb] xtriage to double-check symmetry

Frank von Delft frank.vondelft at sgc.ox.ac.uk
Fri Sep 18 00:30:16 PDT 2009

>>>> So I thought I'd listen to xtriage:  seemed easy since I have both 
>>>> data
>>>> and model in P6522, and I ran:
>>>>     phenix.xtriage final.mtz reference.structure.file=final.pdb
Uh-oh....  what I meant is, I have both data and model in ==>P65<==.  
(Bad typo -- explains Peter's patient but puzzled response, which did 
however confirm I'd understood correctly after all :) 

So, again:  I have *some* lower symmetry (not P1, though);  should I 
expect xtriage to print out tests for P6522 somewhere?  Because I don't 
find it.  

<later> That said:  after updating phenix to 1.4-162, I now do find that 
table in the output (attached, last table), and it seems we're 
in-between:  seems we have probably indeed missed the two-fold, so now 
we must figure out why it "didn't refine" in P6522.

Thanks for you help and explanation!

> But isn't P6522 already the highest symmetry? I think if you run X-triage
> on the data reduced as P65, it will as you say check the extra operations
> implied by p6522 and see if they are justified.
> If the data is already reduced in p6522, those reflections have already
> been merged and there is no way to tell how good the agreement was?
> I don't think xtriage will look at your model- but then I've never 
> given it a model.
> I would guess from your lower Rfree in P65 that it is the lower symmetry-
> the 2-fold operators that make it almost fit p6522 are psuedosymmetry,
> noncrystallographic symmetry that almost fits the p6522 operators but 
> not quite.
> Best,
> Ed
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: xtriage.log
URL: <http://phenix-online.org/pipermail/phenixbb/attachments/20090918/716e0d4e/attachment-0001.el>

More information about the phenixbb mailing list