[phenixbb] Group ADP refinement restraints

[Pavel Afonine]

Hi Taylor,

you are correct, in group B factor refinement overall B factors of 
adjacent residues are restrained to be similar (or not too different). I 
will fix the documentation, thanks for pointing this out.

There isn't much validation done to B factors, so dropping restraints 
can indeed lower R-factors but it isn't necessarily clear whether 
refined B factors are going to be meaningful. I guess you can try both 
and see what produces more meaningful B factors.

Pavel

On 11/25/19 16:34, Taylor Stevens wrote:
> Hello,
>
> I am currently refining a low-ish resolution structure (through the 
> phenix.refine GUI) with phenix version 1.14 and had some questions 
> about the use of restraints in group ADP refinement. In the 
> phenix.refine documentation 
> (https://www.phenix-online.org/documentation/reference/refinement.html#refinement-of-atomic-displacement-parameters-adp-or-b-factors) 
> it is stated that “Group isotropic ADP or TLS refinement do not use 
> any restraints”, however, this does not seem to be the case. If I look 
> in the log file I can see the following lines of text which I believe 
> indicate that the restraints are being applied:
>
> |-group b-factor refinement (macro cycle = 0; iterations = 0)————————|
> | r_work = 0.3331 r_free = 0.3717 target = 5.823062 restraints weight 
> = None  |
> |——————————————————————————————————————|
> |-group b-factor refinement (macro cycle = 1; iterations = 22)————————|
> | r_work = 0.3331 r_free = 0.3717 target = 5.823062 restraints weight 
> = 821.871|
> |——————————————————————————————————————|
> |-group b-factor refinement (macro cycle = 2; iterations = 22)————————|
> | r_work = 0.3331 r_free = 0.3717 target = 5.823062 restraints weight 
> = 821.871|
> |——————————————————————————————————————|
> |-group b-factor refinement (macro cycle = 3; iterations = 22)————————|
> | r_work = 0.3331 r_free = 0.3717 target = 5.823062 restraints weight 
> = 821.871|
> |——————————————————————————————————————|
>
> If I manually set the wu parameter to 0 under “Refinement target 
> weights" nothing changes, however, if I navigate to "All parameters > 
> ADP settings > Group B-Factors" in the GUI there is a “Use restraints” 
> box that is checked by default. If I uncheck this box, there are no 
> longer any restraints being applied according to the log file, and in 
> the case of the structure I am refining, there is a substantial drop 
> in the rfree/rwork. Is this an issue with the program or 
> documentation? Or am I misunderstanding how the program is supposed to 
> work? Is it generally a good idea to use group ADP refinement without 
> restraints?
>
> Thanks,
>
> Taylor
>
> _______________________________________________
> phenixbb mailing list
> phenixbb at phenix-online.org
> http://phenix-online.org/mailman/listinfo/phenixbb
> Unsubscribe: phenixbb-leave at phenix-online.org

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://phenix-online.org/pipermail/phenixbb/attachments/20191126/f61820a3/attachment.htm>