[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [ccp4bb]: Query - bgroup Vs bindividual refinement??



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

raji wrote:
> Is there a value for the data-to-parameter ratio to figure out when one can do
> individual B-factor refinement ? Also, what are the other deciding factors to
> find out when one can switch to individual B-factor refinement ?
> Could you pl. share your suggestions with me.
> Thanks a ton.
> Raji

Hi,

as usual when you want to increase the number of parameters during
refinement, you should check that this leads to an Rfree improvement. If
R improves, but Rfree doesn't change, or even worse, increases, this
means that you are not improving your model by adding these refinable
parameters. Keep in mind that you can always lower you R factor by
increasing the numbers of refinable parameters, but you cannot fudge the
Rfree criteria (IF you rfree set is really independant... take care if
you have NCS).

hope this helps

laurent


-- 


-------------------------------------------------------------------
Laurent Maveyraud               
Groupe de Biophysique Structurale  -  IPBS-CNRS

205, route de Narbonne       	 Tel (+33) 05 61 17 54 35
31077 TOULOUSE CEDEX        	 Fax (+33) 05 61 17 54 48
France                       	 mailto:laurent.maveyraud@ipbs.fr
-------------------------------------------------------------------