[phenixbb] refinement of ligands and waters

Pavel Afonine pafonine at lbl.gov
Wed Jun 10 05:07:27 PDT 2015

Hi Alexandra,

> 1) I have a metal coordinated to a cofactor in the active center of my 
> protein structure. The initial solution provided by 
> phaser/autobuilding already contained the metal+cofactor modelled in 
> the density but with zero occupancy. Then I created all the necessary 
> restraints and cif files with phenix_ready_set and did two cycles of 
> model building/phenix refinement. The refined structures always 
> contained zero occupancy for those atoms. My question is: should I 
> manually change the occupancy for the metal+cofactor to 1? I thought 
> this would be made automatically by phenix because I selected the 
> option “occupancies” and “individual B-factors” in the refinement 
> strategy…

this is explained in the manual:

Please let me know if it is still not clear.

In a nutshell, if you want occupancy to be refined automatically it 
needs to be between 0 and 1 in input PDB file.

> 2) When manually placing waters in the density with coot, what is the 
> maximum value of B-factor that is acceptable for a water? I am asking 
> that because I have some water molecules with weak density and 
> consequently B-factors between 45 and 55 but that are involved in 
> interactions with other molecules. Is it better to keep those waters 
> or remove them?

No need to do it manually. Let phenix.refine update water 
(add/remove/refine) and check them once or so at the end of refinement 
to make sure all water make sense.

There are way more considerations that go into decision-making about 
placing a water than just B-factor; typically they are considered all 


More information about the phenixbb mailing list