I want to about making a constraint file for enzyme design. In that the atom types have to be given according to what we have in the pdb file or we have to use rosetta atom types . If we have to use rosetta atom types how do I have to use it ??
I am 99% certain that all constraint file types use PDB nomenclature for atoms (this is the first atoms column in the ligand's .params file). This is partially because only PDB atom names are unique. Rosetta and MM atom types are only used internally. I'll get someone to doublecheck that but assume it's right if no one corrects me.
As mentioned in the enzdes/match constraint file documentation (http://www.rosettacommons.org/manuals/archive/rosetta3.3_user_guide/app_...) you can give either atom names *OR* atom types in an enzdes constraint file.
Atom names are self explanitory, and are the names found in the atom name spot in the PDB file. (Technically, the ones in the atom name spot in the residue's params file, but since the atom name in the PDB needs to match the one in the params file to be recognized, the difference is insignificant.)
You can also use atom types if you want a more general constraint. For example, have a constraint against a protein carboxylate group, but don't care if it would be a Glu or an Asp which is making contact. In those cases, you would use the Rosetta atom type for the carboxylate oxygen, i.e. OOC. You couldn't use the atom name, as either OD1, OD2, OE1, or OE2 could be the name of the atom making contact. You can find the atom type by looking at the third column of the ATOM lines in the params files (param files for standard amino acids are in rosetta_database/chemical/residue_type_sets/fa_standard/residue_types/l-caa/)
The molecular mechanics types (fourth column in params files) are not used by current enzyme design protocols.