Re: [ccp4bb]: arp warp bad data warning

["Eleanor J. Dodson" <ccp4@ysbl.york.ac.uk>]

***  For details on how to be removed from this list visit the  ***
***          CCP4 home page http://www.ccp4.ac.uk         ***



Maneesh Yadav wrote:
> ***  For details on how to be removed from this list visit the  ***
> ***          CCP4 home page http://www.ccp4.ac.uk         ***
> 
> Hi all,
> I have been having a lot of trouble refining a crystal structure, so I
> tried to use arp warp to add the waters and do the final refinement.
> My R,Rfree value has been stuck at the about 30%, 33% (before solvent).
> The assymetric unit is half of a C2 symmetrical tetrameric 4 helix
> bundle (the C2 axis is along the hydrophobic core).  The sg is P4132 and
> unit cell dimensions are 79x79x79. The resolution of the data is 2.07 A.
> There are no apparent problems during processing of images.


> 
> I've tried lowering the symmetry and the general feel I get from looking
> at other messages is that at these statistics, you can lower the symmetry
> but you are increasing the overall number of parameters per observation,
> even with collecting more data; so lowering is just as bad as any other
> choice you have in order to try lowering the R-factor.
> 
> Arp warp will bring down the values to 25%, 31%; but it complains loudly
> if I try to use the bulk solvent.  My data *seem* fine, and the wilson
> plot looks like that of a standard protein (linear up to 4A and then a dip
> between 5.5-7.0 A).  The chi sq vs. resolution ( I don't really know
> understand this statistic, is it the deriviative of the Wilson in
> someway?)  has a large spike at 6.0A (as it seems to have for every data
> set I have, across many different crystals/days of collection).  Arp warp
> seems to be very sure of itself in telling me that
> my low resolution data is bad and I should cutoff at 3.25A (!).  Is it the
> honest truth, or is there something I should know about the usage of arp
> warp?
> 
> Does anyone have any commments?
> 
> 


Only that arp/warp always complains, whether the data is good or not..
Look at the truncate Wilson plot - if it looks OK I would not worry too 
much.

  Eleanor

> 
>