[phenixbb] unrestrained refinement

Pavel Afonine pafonine at lbl.gov
Mon Jan 28 16:58:22 PST 2019


Hi Jens,

I re-enabled this functionality so that phenix.refine can continue if 
there is an unknown entity found in input model file.
To run unrestrained refinement you can simply use "wc=0" which sets zero 
weight on geometry restraints (this is unrelated to above).

A next available nightly build should have it:

https://www.phenix-online.org/download/nightly_builds.cgi?show_all=1

Let me know of any issues!

Good luck,
Pavel

On 12/18/18 14:58, Kaiser, Jens T. wrote:
>
> Hi!
>
>   I'm trying to do unrestrained refinement of a small molecule in 
> phenix. I have done this many years ago, but it seems whatever I try, 
> phenix.refine now complains about missing nonbonded energy type for 
> all atoms.
>
>  The pdb file I supply has the following format (just one line)
>
>
> HETATM   33  C81 UNK     1       3.117  14.425  18.765 1.00  
> 1.45           C
>
>
> I tried unchecking all references to geometry restraints in the GUI, 
> setting wc and wu to 0
>
>   I also started from a default .eff file and set all references to 
> restraints to either True of False (depending what they said).
>
>   I also tried the line from the phenix.refine documentation:
>
> phenix.refine ../cif_in/mkt_a.ent 
> ../data/181119/proc/137C3/XDS_17000.HKL strategy=individual_sites wc=0 
> refinement.input.symmetry_safety_check=warning 
> refinement.input.xray_data.r_free_flags.generate=True
>
>
>  I'm using phenix v 1.14-3260 from an sbgrid environment.
>
>
>   Any pointers welcome.
>
>
> Cheers,
>
>
> Jens
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://phenix-online.org/pipermail/phenixbb/attachments/20190128/9e727085/attachment.htm>


More information about the phenixbb mailing list