Dear Jacob,
late changes don't compromise the freedom of the free set provided that
you refine roughly to convergence.
The latest reference for this statement can be found in
http://www.pnas.org/content/early/2015/07/02/1502136112
with further references to previous publications on the topic from when
Rfree was introduced by Axel Brunger.
Cheers,
Tim
On 07/23/2015 03:03 AM, Keller, Jacob wrote:
>> If you have twinned data then your Rfree set is already
>> compromised, because pairs of reflections related by the twin law
>> contribute to
> each other's measured intensity. The operation described is an
> obvious way to recover from this state.
>
> Wouldn't the more rigorous way be to re-solve the structure with an
> appropriately-selected free set? Or even better, to assign by default
> twin-proofed free sets from the outset in scaling, as I have heard is
> done in Phenix? Or maybe late changes in free sets are admittedly not
> ideal, but don't end up mattering that much in practice? Has someone
> looked at this? Maybe the structure "forgets" what it had seen in the
> re-assigned free set after a while?
>
> JPK
>
--
--
Dr Tim Gruene
Institut fuer anorganische Chemie
Tammannstr. 4
D-37077 Goettingen
phone: +49 (0)551 39 22149
GPG Key ID = A46BEE1A
|