I am docking an antibody that has 20 residues in the H3 loop. I realize that this is a very challenging case. From the 2017 paper by Weitzner and Gray, I understand the importance of a kink constraint in H3 loop modeling. I want to know if the H3 loop remodeling that occurs during snugdock includes the kink constraint. If not, I am concerned that the kink will be lost during remodeling, especially for this long loop.
In SnugDock, the kink constraint is possible if you add “-antibody:auto_generate_kink_constraint and -antibody:all_atom_mode_kink_constraint” to your options. This should be identical to Rosetta Antibody. You’ll know that it has run if there are “atom_pair” scores in the output score file.
I have been using Snugdock on the Rosie server. Does that have the kink constraint enabled? I suppose it is not enabled, as there is no “atom_pair” score in the output file. Just want to be sure about it.
.
