[phenixbb] phenix.refine

Nathaniel Echols nechols at lbl.gov
Mon May 14 13:05:49 PDT 2012


On Mon, May 14, 2012 at 11:13 AM, Nora Cronin <Nora.Cronin at icr.ac.uk> wrote:
> A refine input file with strategy selecting individual adp refinement has a
> selection of excluded residues as follows:
>
>   adp {
>       individual {
>         isotropic = not ((chain A and resseq 1500:1501) or (chain B and
> resseq 1500:1501) or (chain X))
>         anisotropic = none
>
> The isotropic B factors of the above excluded residues  are however refined.
>   The selection has been ignored.

I believe this is because of the overall B-factor for the entire
crystal ("Ucryst") being refined, which will happen no matter what.
Because phenix.refine reports the total atomic B-factors including all
contributions, this change will propagate to the atoms you've
excluded, even if their individual B-factors aren't being refined.
(Although it is also unclear why you wouldn't want to refine those
B-factors too - this is almost always recommended.)  I guess we need
to add this to the phenix.refine FAQ list.

PS. please post with a new subject instead of replying to a previous
post, it makes a big difference for some email clients and our mailing
list archives.

-Nat


More information about the phenixbb mailing list