We make our monomers as mbuild compounds, then save them as sdf files to create an openff Molecule to feed into espaloma. The atoms in our mol2 files are renamed to correspond to the espaloma atom type. This is solely dependent on the indices of the mbuild compound and the indices of the openff molecule. These indices do not remain the same (i.e. the atoms aren't indexed in the same order) between monomer files and dimer files. To combat this we added a monomer and dimer to the same file and parameterized both at the same time, then took the monomer to create a typed polymer for simulations. This works on most the monomers, but the larger monomers like PIDTCPDT-C11-BO and PIDTFBT-C11-BO do not work.
We need to either find a better way to type the atoms so they're consistent between dimers and monomers, or figure out how to create an openff molecule from a mol2 file.
We make our monomers as mbuild compounds, then save them as sdf files to create an openff Molecule to feed into espaloma. The atoms in our mol2 files are renamed to correspond to the espaloma atom type. This is solely dependent on the indices of the mbuild compound and the indices of the openff molecule. These indices do not remain the same (i.e. the atoms aren't indexed in the same order) between monomer files and dimer files. To combat this we added a monomer and dimer to the same file and parameterized both at the same time, then took the monomer to create a typed polymer for simulations. This works on most the monomers, but the larger monomers like PIDTCPDT-C11-BO and PIDTFBT-C11-BO do not work.
We need to either find a better way to type the atoms so they're consistent between dimers and monomers, or figure out how to create an openff molecule from a mol2 file.