I am trying to add riding hydrogens to a model that includes beta-3-homoglutamine (B3Q). The monomer is not in the ccp4 library but it is a PDB chemical component. So far I have used elbow to build and optimize my own version of the ligand using the atom naming convention for a beta-amino acid in the PDB (where CB is in the backbone adjacent to the amino group, so the backbone goes like C-CA-CB-N). I am using a TRANS cif link from the monomer library for the N(alpha-amino acid) to C(beta-amino acid), and I have defined my own cif link for the N(beta-amino acid) to C(alpha-amino acid). However when I run phenix.reduce I can't get hydrogens to appear on the peptide bonds between the beta-amino acid and adjacent alpha-amino acids (hydrogens appear every where else on the ligand and in the model).

I realize the effect of two hydrogens on the model is inconsequential but I would like to figure out how I can get phenix.reduce to recognize the bonds to my "ligand" as standard peptide bonds and add hydrogens accordingly.