I am working on integrating the Backrub Protocol described in Friedland et. al. PLoS Computational Biology 2009 into one of my current scripts in PyRosetta.
The main issue I am encountering is that I would like to know what specific residues (or in this case which residues are present in the particular segment) to which the move was applied. Essentially the published protocol states that rotamers were optimized for a random residues within the segment that was moved by the Backrub Mover, but I am not aware of any method by which I can easily determine which residues those are.
This may depend a bit on how you're doing the backrub, but if you're using the protocols.backrub.BackRubMover, there isn't any sort of "saving" of the residues used. The best you can do is to enable the `-out:levels protocols.backrub.BackrubMover:debug` as a command line option when you intialize Rosetta, and some information about which segments the Backrub protocol chose for manipulation.