[phenixbb] unwanted N/Q/H flipping in PHENIX.REFINE

[Pavel Afonine]

Hi Bernhard,

If you disable the flips they should not happen during refinement.
This is something that I heard before but users were reluctant to 
provide evidence.
Could you please send me data file, and two PDB files (before and after 
refinement) and indicate residues that are undesirably flipped? In case 
there are ligands, please send ligand CIF files too.
Once I have files I will run refinement myself and see how this behavior 
may be prevented.

Thanks,
Pavel


On 12/29/15 09:38, Bernhard Loll wrote:
> Dear all,
>
> I am currently refining a 1.9 A structure with PHENIX 1.10.1-2155.
>
> The problem I am facing is the following:
>
> Running the Molprobity server, it suggests me to correct a couple of N/Q/H
> flips. I perform the flips re-run PHENIX.REFINE with  “Automatically
> correct N/Q/H  errors” not ticked. The log file confirms the selection
> with nqh_flips = False.
>
> Nevertheless PHENIX.REFINE again flips these residues and the output PDB
> file has all previously flipped residues “re-flipped” in comparison to the
> input PDB file. I am very confident that the automatically flips performed
> by PHENIX.REFINE (REDEUCE) are wrong, since I have two metal binding
> sites.
>
> Why PHENIX.REFINE rotates side chains by 180 degree even though nqh_flips
> were set to false?
>
> Any help is appreciated.
>
> Cheers,
>
> Bernhard
>
>
> Dr. Bernhard Loll
> Freie Universitaet Berlin
> Fachbereich Biologie, Chemie, Pharmazie
> Institut fuer Chemie und Biochemie
> AG Strukturbiochemie
> Takustr. 6
> D-14195 Berlin
> Germany
>
> Phone: +49 (0) 30 838-57348
> Fax:   +49 (0) 30 838-454936
> Email: loll at chemie.fu-berlin.de
>
> Homepage: http://www.bcp.fu-berlin.de/chemie/bc/ag/agwahl/
>
> _______________________________________________
> phenixbb mailing list
> phenixbb at phenix-online.org
> http://phenix-online.org/mailman/listinfo/phenixbb
> Unsubscribe: phenixbb-leave at phenix-online.org