Hi Mark, hm... this is strange as I can't remember any relevant changes that would trigger this behavior.. Do you have two PDB files, one from old and one from new Phenix versions that exemplify the problem (preferably with REMARK 3 headers that show Phenix version number)? Could it be the other way around: coot's convention is now different from phenix? Phenix uses GeoStd geometry library which doesn't change much apart from adding new entries. This is rather Nigel's domain so he may comment on this more. However, I would say we need more specific information to investigate this. Pavel

Apparently a torsion angle of –180° is considered very different from 0° because of the NH1.NH2 names. And the phenix convention is now different from coot.

Mark _______________________________________________ Mark A. Saper, Ph.D. Associate Professor of Biological Chemistry, U-M Medical School New location: Room 3220B, MSRB III | [email protected] | +1 (734) 764-3353

On Feb 14, 2016, at 4:07 PM, Eckhard Hofmann wrote:

...

Dear Mark, I think Oleg is in the process of writing a little routine to fix problems in pdbs with regard to order of atoms in residues and such. Maybe fixing naming conventions could/will be included. Cheers, Eckhard

Am 14.02.2016 um 19:37 schrieb Mark A Saper:

...

Phenix 1.10.1 is flagging several of my Arg residues with bad dihedral angles. Phenix is much more strict about NH1, NH2 naming conventions then 6 months ago. Can these be fixed with PDB Tools? How? Many thanks in advance. -Mark