Hi Pavel, yes I do have the file which goes : geometry_restraints { edits { bond { atom_selection_1 = chain A and resseq 1601 and name O2 atom_selection_2 = chain A and resseq 895 and name OG distance_ideal = 3.2 sigma = 0.5 } bond { atom_selection_1 = chain A and resseq 1601 and name O2 atom_selection_2 = chain A and resseq 299 and name OG distance_ideal = 2.3 sigma = 0.5 } .... many more bonds.... } } But it takes some time to generate a good amount of restraints, and for structures with several ligands it could be a long task. Agreed that it is not fool-proof but a general edition of these bonds and editing afterwards would be a good little program I believe. Best Vincent Le 15/09/2023 à 14:34, Pavel Afonine a écrit :
Hi Vincent,
Before launching phenix.varref, I would like to generate a ligand restraint file so the ligand doesn't fly away from its density. It would be great if the ligand_restraints.eff file could be generated from a PDB file already containing the ligand (for example originating from refinement), that we can then edit afterwards. The ligand I'm talking about is ATP in this case, so it doesn't need external cif file for refinement.
I searched but couldn't identify such a job. It would be useful if it could be added to "PDB tools" or similar.
you need to define what in Phenix is called custom bond restraints: some "dummy" bonds between pairs of ligand and protein atoms of your choice. These bonds will anchor the ligand in place. Let me know if you need help setting up those but I think I sent you an example some time ago. While generating such a file automatically is possible it may not be fool-proof as it will depend on the geometric quality of the model. So it's best to use your chemical intuition to choose those pairs of atoms that you'd like to link.
Good luck! Pavel
-- Vincent Chaptal, PhD Director of GdR APPICOM Drug Resistance and Membrane Proteins Lab MMSB -UMR5086 7 passage du Vercors 69007 LYON FRANCE +33 4 37 65 29 01 http://www.appicom.cnrs.fr http://mmsb.cnrs.fr/en/