[phenixbb] Partial occupancy refinement of overlapping inhibitor+water sites

[Eckhard Hofmann]

Hi Matt,
you will have to define the appropriate occupancy groups.
I usually put that into a separate file (e.g. occ.params) which you then 
include as input.
In your case this should do the job (I think  ;-)):

refinement {
   refine {
     occupancies {
       constrained_group {
         selection = (chain A and resseq 802)
         selection = (chain W and resseq 1:6)
       }
     }
   }
}

This should refine both selections to a sum of 1.
More examples are at
https://phenix-online.org/documentation/reference/refinement.html#occupancy-refinement

Cheers,
Eckhard

Am 14.03.24 um 22:27 schrieb Matt Jordan McLeod:
> Hi all,
> 
> I am trying to navigate phenix.refine to occupancy refine a partially 
> occupied inhibitor vs. the waters that it displaces.
> 
> For instance I have molecule resseq 802 in chain A, and waters 1-6 in 
> chain W that are overlapping.  When just refining occupancy of the 
> molecule there are clear sites for the waters.   How can I refine both 
> of them together so that I can get an occupancy of the molecule vs the 
> group of waters (which should have nearly identical occupancies)?
> 
> Maybe, a bit more information on the differences in refinement strategy 
> for the occupancy vs fix occupancy vs group occupancy would be more 
> broadly helpful.
> 
> I tried to make them all a group, but that didn't work as the refinement 
> pushed the waters out and reduced their occupancies to ~0 while leaving 
> the small molecule occupancy about the same as without water.
> 
> **Waters are only placed, they are not refined), small molecule is 
> refined to 0.71 occupancy.*​*
> 
> --
> *Matthew Jordan McLeod, PhD*
> Post-Doctoral Fellow |//Thorne Lab
> Cornell University
> G11 Physical Sciences Building
> 142 Sciences Drive
> Ithaca, NY 14853-3501
> (607) 279 9382
> 
> _______________________________________________
> phenixbb mailing list
> phenixbb at phenix-online.org
> http://phenix-online.org/mailman/listinfo/phenixbb
> Unsubscribe: phenixbb-leave at phenix-online.org

-- 
Prof. Dr. Eckhard Hofmann <eckhard.hofmann at rub.de>
Ruhr-Uni Bochum
AG Proteinkristallographie
ND04/318, Fachnummer 50
44780 Bochum
Tel: +49-(0)234/32-24463, Sekr. -24461, FAX: -14762
ORCID: 0000-0003-4874-372X