You’ll need to provide parameters for the lipid molecules you have in the input molecule. See https://www.rosettacommons.org/demos/latest/tutorials/prepare_ligand/prepare_ligand_tutorial for a tutorial on how to do this.
Alternatively, you may just want to use the automatic generation of parameters. If you add the `-load_PDB_components` option to the intialization of PyRosetta, it can attempt to autogenerate missing ligand parameters from the PDB Chemical Components Dictionary file. This can be downloaded from http://www.wwpdb.org/data/ccd, and either can be placed into the Rosetta database (in database/chemical/), or can be specified by passing the path to the file with the -PDB_components_file option. — Keep in mind, though, that the from-CCD generation is rather simplistic. It sufficies for having the ligand be present, but the molfile_to_params way allows you more control of how the ligand is set up and how it’s sampled.
A final caveat — peptide/lipid interactions are a rather unexplored territory for Rosetta. You’re exploring new ground. Be aware that things might not work right-off-the-bat, and that you’ll need to do a fair amount of testing and benchmarking, and will likely need to adjust/come up with new protocols.
.
