Member Site › Forums › Rosetta 3 › Rosetta 3 – General › positive results in total_score by symmetry docking › Reply To: positive results in total_score by symmetry docking
Positive scores mean that Rosetta doesn’t like the structure very much. However, that doesn’t mean that the docked conformation is bad. Most of the structures you download directly from the PDB have positive scores, as Rosetta’s energy function is highly sensitive to slight variations in position. Small movements that crystallographers might not bother with can have big effects on Rosetta score. It’s recommended to minimize or relax a structure to relieve these energy issues prior to using it in Rosetta. (See http://www.rosettacommons.org/manuals/archive/rosetta3.2.1_user_guide/preparing_structures.html for a number of options. Using the relax application with the -constrain_relax_to_start_coords command line flag is probably the most straightforward.)
My guess is that there’s something about the starting structure that Rosetta doesn’t like. (I’ve seen cases where there is a bad twist in just the two C-terminal residues, and minimizing just those two residues will change the score by 200 points). The monomers probably start off with a positive total_score, so the complex ends up with a positive total_score, although lower than the summed monomers. Usually for docking cases you don’t really want to look at the total_score of the complex, but at the binding score (different applications call this different things) – not the total energy of the complex, but the interaction energy or what Rosetta thinks is equivalent to the ddG of binding. This way you can compensate for differing internal energies of the monomers, or even compare complexes with different sizes of monomers.