[phenixbb] advice on 'Fit rotatable H or D' refinement stage in phenix.refine

Pavel Afonine pafonine at lbl.gov
Wed Jan 7 13:47:33 PST 2015


Hello Anna,

if you want to refine H atoms (coordinates, B-factors) individually then use

hydrogens.refine=individual

Note, this will only make sense if data resolution is high enough, say 
better than 0.9-0.7A or so.

You can disable 'Fit rotatable H or D' by using

hydrogens.real_space_optimize_x_h_orientation=False

Note, this has nothing to do with torsion angle dynamics.

Pavel

On 1/7/15 8:26 AM, Anna Makal wrote:
> Hi,
>
> I would like to ask for advice concerning 'Fit rotatable H or D' 
> refinement stage in phenix.refine.
>
> I am attempting to run unconstrained refinement on a certain 
> high-resolution protein dataset: just XYZ and ADPs refinement for 
> every single atom in the structure, with hydrogen atoms treated with 
> individual approach.
>
> It turns out that when I set more then 3 macrocycles, the 'Fit 
> rotatable H or D' refinement stage is activated in every macrocycle 
> prior to 'XYZ individual' refinement stage.
> I am using version 1.9, release 1692 for windows.
> When I last worked with PHENIX, it was distribution 1342, and I am 
> positive I have not come across such refinement stage. It seems that 
> TARDY is involved at that stage.
>
> As I specifically do not want to introduce any kind of torsion angle 
> dynamics in my current refinement, I would like to ask, what 
> command/option shall I use to deactivate this stage of H-atom fitting 
> in my refinement?
>
> With regards,
>
> Anna Makal
> _______________________________________________
> phenixbb mailing list
> phenixbb at phenix-online.org
> http://phenix-online.org/mailman/listinfo/phenixbb



More information about the phenixbb mailing list