[phenixbb] sigma_iobs_rejection_criterion

Pavel Afonine pafonine at lbl.gov
Tue Jul 10 16:09:11 PDT 2012

Hi Yarrow,

if you run refinement with F, then only F>=0 are used.
If you run refinement with I, then the French&Wilson procedure of 
converting I to F is used:

French, G.S., Wilson, K.S. (1978) On the treatment of negative intensity 
observations. Acta Cryst A34, 517–525.


On 7/10/12 2:38 PM, Yarrow Madrona wrote:
> Hi,
> Do you know where I could find the default values for criteria for error
> in Iobs and Fobs? I am having trouble finding it in the log file. Thanks.
> -Yarrow
>> On Tue, Jul 10, 2012 at 8:58 AM, Yarrow Madrona <amadrona at uci.edu> wrote:
>>> 1. Is this the same observed criterion that is reported in the RCSB
>>> Deposition?
>>> 2. To be honest I don't understand what this is. Is it a limit for the
>>> error in Iobs, below which the reflection will be discarded?
>>> 3. What is a reasonable setting for this input and how does it change
>>> the
>>> refinement?
>> To be honest, this was an incompatibility between the current code and
>> older parameter files that we hadn't anticipated.  In the context of
>> French & Wilson treatment, I believe the parameter basically filters
>> out excessively negative intensities.  I would probably leave this
>> alone - you can get the old parameter file to work again by resetting
>> it to None, which is the new default value.  The current nightly build
>> has been fixed to prevent it from stopping if it sees a value of 0
>> instead.
>> -Nat
>> _______________________________________________
>> phenixbb mailing list
>> phenixbb at phenix-online.org
>> http://phenix-online.org/mailman/listinfo/phenixbb

More information about the phenixbb mailing list