[phenixbb] options in phenix.merging_statistics

Dorothee Liebschner dcliebschner at lbl.gov
Thu Mar 23 16:29:51 PDT 2017


Hi,

It is up to you to choose option 1, depending on what information you want
to get. Do you have anomalous data?

Concerning the difference between "no options" and "option 1", what change
in statistics is bothering you?
It is expected that multiplicity (redundancy) decreases, as Friedel pairs
are considered separate.
Rmerge is (slightly) higher for the no merge option, as both Friedels are
treated as equivalent (--> higher multiplicity --> higher Rmerge) and
internal agreement of the intensities is worse, if anomalous differences
are large.
Similarly, completeness is expected to be sligthly lower if Friedel pairs
are considered separate.
The entries for I/sigma seem to be not according to column headers, so I am
not sure, what value belongs to which option.

If the data is not anomalous, I would not choose option 1 (unless you have
a reason you want to use it).

Best wishes,

Dorothee



On Thu, Mar 23, 2017 at 3:19 PM, Murpholino Peligro <murpholinox at gmail.com>
wrote:

> Hello.
>
> I was wondering what is the normal situation where one should click the
> boxes marked as "Keep anomalous pairs separate in merging statistics"
> (option1) and/or
>  "Use internal variance of the data in the calculation of the merged
> sigmas" (option2)?
>
> I am asking because it looks like it is detrimental to the statistics, as
> you can see in the file attached (CSV file with 4 columns for no options,
> option1, option2 and option1&2).
>
> Thanks in advance.
>
>
> MP
>
> _______________________________________________
> phenixbb mailing list
> phenixbb at phenix-online.org
> http://phenix-online.org/mailman/listinfo/phenixbb
> Unsubscribe: phenixbb-leave at phenix-online.org
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://phenix-online.org/pipermail/phenixbb/attachments/20170323/83a0a3b0/attachment-0001.htm>


More information about the phenixbb mailing list