Hi Jarrod,
If this occurs in a not yet refined model, this is not a problem. It is an indication that your test set is free yet. If Rfree stays lower than R at the later stages of model building and refinement this typically means that your test set is not independent, that is, test set reflections have closely related reflections in the work set. This can be caused by NCS, twinning and using a space group of lower symmetry than the real spacegroup.
Cheers,
Robbie
> -----Original Message-----
> From: CCP4 bulletin board [mailto:[log in to unmask]] On Behalf Of
> Jarrod Mousa
> Sent: Saturday, January 24, 2015 19:05
> To: [log in to unmask]
> Subject: [ccp4bb] Rfree<Rwork
>
> Hi all,
>
> I am wondering what the conditions are when Rfree is less than Rwork. I
> have seen this when trying certain models to solve my structure, but I am not
> sure why this occurs, or if this is valid or not.
>
> Thanks,
>
> Jarrod Mousa
>
> U. of Florida
|