Reply To: SequenceRelax on a loop

[Anonymous]

The scorefunction term is chainbreak, or something similar to that (maybe cbreak?). Looking at your code, it seems likely it’s not turned on. If you can’t find it I’ll look for it in the C++.

set_single_loop_fold_tree(pose, loop)

This is the real heart of what the chain break does and why it’s important. The loop is kinematically isolated from the rest of the structure by routing protein folding AROUND the loop using a fixed jump between the loop’s termini. Because the fold tree must form a directed acyclic graph, and the termini are connected, a break is necessary. If you don’t isolate and break in this fashion, then when you change loop torsions, the entire downstream half of the protein moves with the loop.

Skipping the loop fold tree call is acceptable if you do only minimization, because the minimizer will penalize breaking the protein in two, but you will see some core shifts if you do this.

Read Chu’s paper about the fold tree for some more details (um, I think).

J Mol Biol. 2007 Oct 19;373(2):503-19. Epub 2007 Aug 2.
Protein-protein docking with backbone flexibility.

Wang C, Bradley P, Baker D.