Hi Cory, Did you supply phenix with a "edits" file that indicates that there is a glycosidic bond between the Thr and the NGA? Such file is a simple text file with content like this: refinement.geometry_restraints.edits { bond { action = *add atom_selection_1 = name C1 and chain A and resname NGA and resseq 1001 atom_selection_2 = name OG1 and chain A and resname THR and resseq 100 distance_ideal = 1.439 sigma = 0.050 } } The above generates a generic covalent bond between C1 of NGA and OG1 of THR. Alternatively, you can use the NAG-THR link that is already in the standard monomer library. (Note that this is actually for GlcNAc-Thr, but it should work for NGA-Thr too. Yes, its a hack.) This way you get a few more restraints: refinement.pdb_interpretation.apply_cif_link { data_link = NAG-THR residue_selection_1 = chain A and resname NGA and resseq 1001 residue_selection_2 = chain A and resname THR and resseq 100 } Note the NAG and NGA. For your reference, the NAG-THR link is defined in the monomer library as: ###### NAG-THR . DEL-O1 pyranose THR DEL-HG1 . bond_NAG-C1_=_THR-OG1 ###### and ###### data_link_NAG-THR # loop_ _chem_link_bond.link_id _chem_link_bond.atom_1_comp_id _chem_link_bond.atom_id_1 _chem_link_bond.atom_2_comp_id _chem_link_bond.atom_id_2 _chem_link_bond.type _chem_link_bond.value_dist _chem_link_bond.value_dist_esd NAG-THR 1 C1 2 OG1 single 1.439 .020 loop_ _chem_link_angle.link_id _chem_link_angle.atom_1_comp_id _chem_link_angle.atom_id_1 _chem_link_angle.atom_2_comp_id _chem_link_angle.atom_id_2 _chem_link_angle.atom_3_comp_id _chem_link_angle.atom_id_3 _chem_link_angle.value_angle _chem_link_angle.value_angle_esd NAG-THR 1 C1 2 OG1 2 CB 108.700 3.000 NAG-THR 1 O5 1 C1 2 OG1 112.300 3.000 ##### Good luck. Zhijie -----Original Message----- From: Cory Brooks Sent: Monday, April 13, 2015 3:13 PM To: [email protected] Subject: [phenixbb] Link NGA and THR Hello! I am refining a structure with O-link glycosylation (GalNAc linked to Thr). Phenix seems to remove the link during refinement. Can anyone suggest a solution that will maintain the bond during refinement? Regards, Cory Cory Brooks,Ph.D. _______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb Unsubscribe: [email protected]