selectively remove bond length restraints on atoms
hello, What is the best way to selectively remove bond length restraints on some selected atoms for phenix.refine? i found the commands for a .param file: refinement { geometry_restraints.remove { angles = None dihedrals = None chiralities = None planarities = None } } but no bond length command in the manual online. http://www.phenix-online.org/version_docs/dev-283/refinement.htm otherwise there is the commands refinement { geometry_restraints.edit { bond { action= *delete change atom_selection_1= (selection1 here) atom_selection_2= (selection2 here) distance_ideal= None } } } is it a good way to deal with the problem ? thanks in advance for your help. Maxime
Sorry this isn't supported right now. I'm currently working on the
restraints system, so there is hope that it will work eventually.
Ralf
On Thu, Mar 15, 2012 at 8:20 AM, Maxime Cuypers
hello,
What is the best way to selectively remove bond length restraints on some selected atoms for phenix.refine? i found the commands for a .param file:
refinement { geometry_restraints.remove { angles = None dihedrals = None chiralities = None planarities = None } }
but no bond length command in the manual online. http://www.phenix-online.org/version_docs/dev-283/refinement.htm otherwise there is the commands
refinement { geometry_restraints.edit { bond { action= *delete change atom_selection_1= (selection1 here) atom_selection_2= (selection2 here) distance_ideal= None } } }
is it a good way to deal with the problem ? thanks in advance for your help.
Maxime
_______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb
On Thu, Mar 15, 2012 at 8:20 AM, Maxime Cuypers
What is the best way to selectively remove bond length restraints on some selected atoms for phenix.refine?
This isn't a perfect solution, but I recently added an option to change the weight on selected geometry restraints: refinement.geometry_restraints.edits { scale_restraints { atom_selection = chain A and resseq 1000 and name MG scale = 0.0 apply_to = *bond } } The intended purpose was actually to increase the geometry weight on problematic regions of the structure, but the reverse should be possible. Note however that this will effectively cancel out *all* bond restraints for the selection, not a specific linkage. -Nat
participants (3)
-
Maxime Cuypers -
Nathaniel Echols -
Ralf Grosse-Kunstleve