I am using Rosetta 3.3 and I have a problem with matching. In my cstfile I want to model a H-bond using the atom_type: Hpol of several residues. Results from Matcher are completely wrong in the case of Tyr. Hpol atom is not located on the phenolic oxygen but on the CD1 atom of Tyr which is now double protonated. I was looking into ~/rosetta3.3/database/chemical/residue_type_sets/fa_standard/residue_types/l-caa but Tyr is here defined correctly. I tried to delete and create Dunbrack02.lib.bin once again but no change was observed.
Do you have any suggestion what can be possibly wrong?
that’s certainly a very weird error you’re describing, I have no clue how this could come about. would you mind sending me your testcase (cstfile / scaffold setup / ligand .params file) so that i can look into it? also, what happens when you try to generate a model of the theozyme with the CstfileToTheozymePDB app?
one general comment: when matching hydrogen bonds, I prefer to use the heavy atoms that are involved instead of the protons. that makes it easier to specify a meaningful dihedral around the donor group.