[phenixbb] query re local grid search in phenix.real_space_refine

Pavel Afonine pafonine at lbl.gov
Sat Feb 10 10:41:50 PST 2018

Hi Oliver,

no, it does not do it for all residues. It does it for residues that 
have poor fit to the map or unlikely rotameric state. There is a 
potential to make this smarter and faster.


On 2/10/18 09:41, Oliver Clarke wrote:
> Related to the last message - when using local_grid_search, does phenix attempt to torsion-angle search every residue, or just outliers?
> It would seem like maybe using EMRinger or similar to identify residues where there is strong sidechain density, but in a non-rotameric position or with an incorrectly fit rotamer might be one way to make it more efficient if it is the former.
> I don’t want phenix trying to do a local grid search on residues with no sidechain density - these may be in an area of the map with poor local resolution, perhaps where I have docked a crystal structure, or carefully manually adjusted rotamers - local grid search in these regions is likely to make things worse if it is not focused on residues with clear sidechain density.
> Cheers
> Oli
> _______________________________________________
> phenixbb mailing list
> phenixbb at phenix-online.org
> http://phenix-online.org/mailman/listinfo/phenixbb
> Unsubscribe: phenixbb-leave at phenix-online.org

More information about the phenixbb mailing list