[phenixbb] high resolution bin statistics

[Pavel Afonine]

Hi Sebastiano,

the number of bins used is such that each bin has enough amount of 
"free" reflections for meaningful statistics calculation. For example, 
in your case below only 4 bins allow slightly more than 100 "free" 
reflections per bin. Making more bins will further reduce the number of 
"free" reflections per bin making it impossible to compute Rfree in each 
bin.

Pavel.


On 8/31/09 3:21 AM, Sebastiano Pasqualato wrote:
> Hi there,
> i just wanted to know how is determined the number of bins for which  
> the statistics of refinement are given in the header of the pdb  
> refined with phenix.refine, and if there's a way to have more than the  
> number of bins that is default output.
> For example, what if I'd like to have statistics for the 2.6-2.5 bin,  
> here:
>
> REMARK   3  FIT TO DATA USED IN REFINEMENT (IN BINS).
> REMARK   3   BIN  RESOLUTION RANGE  COMPL.    NWORK NFREE   RWORK  RFREE
> REMARK   3     1 32.4159 -  3.9625    0.99     2545   131  0.1921 0.2232
> REMARK   3     2  3.9625 -  3.1460    1.00     2447   116  0.1650 0.2289
> REMARK   3     3  3.1460 -  2.7486    1.00     2433   121  0.1917 0.2682
> REMARK   3     4  2.7486 -  2.4974    1.00     2405   130  0.2244 0.2890
>
> thanks a lot in advance,
> ciao
> s
>
>