RE: [ccp4bb]: ccp4bb- Warp-Arp

[Bart Hazes <bhazes@ualberta.ca>]

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


For ARP/wARP the relevant quantity is not so much Rfree, but the ability to
derive meaningful SIGMAA coefficients for the maximum likelihood target. Will
the 1/sqrt(n) relation apply to that as well. If so we need to know what
accuracy we need (can get away with). If 5% is tolerable then just 400
reflections would do the job. Perhaps the fact that Sigmaa has to be estimated
as function of resolution increases the number of rfree reflections. Randy,
Raj, Garib ...???

Bart

On Tue, 4 Feb 2003, Bernhard Rupp wrote:

> ***  For details on how to be removed from this list visit the  ***
> ***          CCP4 home page http://www.ccp4.ac.uk         ***
>
> Tassos wrote:
>
> > A small note: I always use 1000-1500 reflections for Rfree instead
> > of the more 'standard' 10%. Saving on a few reflections sometimes
> > for me simply does the trick.
>
> This 10% indeed is the same unreflected number like 10mg/ml conc
> for crystallization. Meth enz 277 B pp 382 essentially states
> in more cryptic words that 'as many reflections as necessary to make
> free R statistically meaningful' i.e., the error in Rfree reasonable.
> That std error (p384) appears to be Rfree/sqrt(n), with n the number of
> test
> reflections, and not a percentage.
>
> Handy ref in case one needs to respond to a review comment.
>
> BR
>
>
>

===============================================================================

Assistant Professor
Dept. of Medical Microbiology & Immunology
University of Alberta
1-15 Medical Sciences Building
Edmonton, Alberta, T6G 2H7, Canada
phone:	1-780-492-0042
fax:	1-780-492-7521

===============================================================================