U is an RNA nucleic acid. Non-RNA Rosetta executeables recognize only DNA by default, because the DNA and RNA names conflict, and the authors decided to offer one-or-the-other instead of figuring out a system for both. Are you doing RNA-protein complexes or DNA-protein complexes?
If this is not the problem it’s probably a nomenclature thing. Your residue names need to match what Rosetta has in its parameter files in the database (database/chemical/residue_type_sets/fa_standard/ or something like that). Note that the RNA is NOT in the fa_standard set, but DNA is. If you’re using DNA I can help you debug this path.