There are many different ways of representing molecular data (InChi, SMILES, CAS, mol, xyz, pdb, etc.). Which situation suites each data type the best (e.g. .pdb for protein data storage, InChi for database representation)? Which one do you prefer and why?