Re: [ccp4bb]: Refmac vs. cns

[Garib Murshudov <garib@ysbl.york.ac.uk>]

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

Usually good indicator is behaviour. I.e. difference between initial and
final values. As Eleanor noted scalings
are different.

Garib

"Eleanor J. Dodson" wrote:

> ***  For details on how to be removed from this list visit the  ***
> ***          CCP4 home page http://www.ccp4.ac.uk         ***
>
> Mohamed Rafi wrote:
> >
> > ***  For details on how to be removed from this list visit the  ***
> > ***          CCP4 home page http://www.ccp4.ac.uk         ***
> >
> > Dear all:
> > May be this is an oldie but..
> > I refined my structure using Refmac5 and CNS using the same set of Rtest
> > reflections. Always, Refmac5 gave a lower R-factor compared to CNS using
> > max likelyhood refinement. Has anybody else noticed this? Why would this
> > occur?
> > Thanks
> > Rafi
>
>  I am sure we would like to say that this was because refmac is better -
> and of course it is.. BUT the R factor within a few decimal places  you
> get is very much a function of your scaling algorithm and at low
> resolution the two can differ quite a lot....
>
>  Just for comparison - do you have resolution? did you use bulk solvent?
> TLS? etc etc..

--
------------------------------------------------------------------
Garib N. Murshudov, Chemistry Department, University of York, U.K.
Tel: Home +44 (1904) 42 62 87, work:  +44 (1904) 43 25 65
------------------------------------------------------------------