Dear Valérie, Can you please share the model and reference model (off-list) that you are using? Then I'll be able to tell exactly what is happening (files will be treated confidentially). For some reason Phenix does not believe they are close enough. That could be because sequence is not similar enough. Renumbering the residues normally won't change this fact. Reference chains needs to be at least 0.85 sequence identical by default. This can be changed with reference_model.search_options.chain_similarity_threshold=0.5 Best regards, Oleg Sobolev. On Wed, Apr 29, 2026 at 12:11 AM Valérie Biou via phenixbb wrote:

Dear all,

I would like to use a reference model to try to improve my cryo-em refinement geometry. I have an AF3 model that superposes with an RMSD of 1.4Å wrt my structure, and I renumbered the residues so they match with my structure. however, I get the message

Reference Model Matching Summary: reference file: renumbered.pdb Model: Reference: Total # of matched residue pairs: 0 Total # of reference model restraints: 0

input file list (truncated from the root) default_model = "RealSpaceRefine_72/rsr71_isolde1_real_space_refined_072.pdb" real_map_files = /cryosparc_P29_J343_010_volume_map_sharp.mrc" default_real_map = cryosparc_P29_J343_010_volume_map_sharp.mrc" restraint_files = eLBOW_46/elbow.P.P_pdb.001.cif" default_restraint = eLBOW_46/elbow.P.P_pdb.001.cif" phil_files = Phenix/.phenix/project_data/real_space_refine_88.eff" default_phil = Phenix/.phenix/project_data/real_space_refine_88.eff"

then : reference_model { enabled = True file = "renumbered.pdb" side_chain = False secondary_structure_only = True search_options { chain_max_rmsd = 3.0

Can you please help me with this? Kind regards

Valérie _______________________________________________ phenixbb mailing list -- [email protected] To unsubscribe send an email to [email protected] Unsubscribe: phenixbb-leave@%(host_name)s