Author: Ingemar André
This document was edited Aug 20th 2010 by Ingemar Andre. This application in rosetta was created by Ingemar André.
The code for the symmetric docking application is in rosetta/main/source/src/apps/pilot/andre/SymDock.cc. See rosetta/tests/integration/tests/symmetric_docking for an example of symmetric docking protocol and input files.
Andre I, Bradley P, Wang C, Baker D. Prediction of the structure of symmetrical protein assemblies. Proc Natl Acad Sci U S A. 2007 Nov 6;104(45):17656-61.
The application predicts the structure of symmetric homooligomeric protein assemblies starting from the structure of a single subunit. It is very similar in spirit to the standard protein-protein docking protocol and contains the same basic components. The relative subunit orientation and conformations of side-chains are simultaneously optimized for a symmetric protein assembly.
The protocol starts by generating an initial symmetric assembly configuration described by the symmetrical definition given as a input to rosetta. The rigid body position (the relative orientation between subunits) is then randomized to get a unbiased starting state. The protocol proceeds with a low resolution stage where the conformational universe is reduced to only configurations in which the subunits form significant amount contacts between them. First the subunits are translated along the allowed degrees of freedom into glancing contact. In a symmetric system there may be several translation degrees of freedom and a multidimensional slide might be required. In the simplest and most common type of symmetry, twofold symmetry (c2), translation can only occur along an axis that connects the the center of mass of the two subunits. This is typically set up to be the x-axis of the system and translation is only allowed along x. With more complicated symmetries there will often be several directions that translation can happen while maintaining the overall symmetry of the system. Generally, how slide moves are performed can be controlled by a parameter in the symmetry definition input. By default, a slide degree of freedom is randomly selected and subunits are translated along this direction until contact is found with other subunits. Then the next translational degree of freedom is selected and a slide into contact is performed. The next step is to optimize the rigid body orientation using Monte Carlo search using a low resolution energy function (same as docking, but symmetric).
The protocol proceeds with a high-resolution stage that simultaneously optimizes the side-chain and rigid body conformation using a Monte-Carlo Minimization (MCM) strategy. This mimics the standard heterodimeric docking protocol.
Protein assemblies can adopt multiple types of symmetrical assemblies. Rosetta can model any type of symmetry. You can find a short description of the more common symmetry types in the Symmetry User's Guide. For a more thorough description of symmetry in biological macromolecules see "Structural symmetry and protein function" Goodsell DS and Olsson AJ, Annu Rev Biophys Biomol Struct. 2000;29:105-53.
Observe that for more complex symmetries you may have to edit the definition file to select in which order translation (sliding-into-contact) movements are supposed to be done. Read the Symmetry User's Guide. for reference.
-database [file path] - the Rosetta 3 database location
-in:file:s [file path] - the input subunit structure
-ex1, -ex2, -extrachi_cutoff [num residues], etc. - increase the resolution of the rotamer library
Currently the best place to read about packing options in general is in the documentation for the fixed backbone design application .
Note, that some options does not make sense in the context of symmetry (such as randomize1 or randomize2, spin, and that dock_pert is covered by the -symmetry:perturb_rigid_body_dofs flag (below) instead etc).
Similar to regular docking it generally makes sense to start with a prepacked protein monomer.
Because of the fewer degrees of freedom in a symmetric system it is generally not required to run as many models as in regular docking. For a cyclical system 5000 models should be enough. For many systems it is not necessary to explicitly simulate all subunits in the symmetrical systems. for example in a dodecamer with cyclical symmetry a subunit only interacts with neighbor subunits to the left or right of it in the ring. With the make_symmdef_file_denovo you can add the -subsystem option to generate a smaller symmetry definition for a smaller system. With the make_symmdef_file script this can be controlled by finetuning the cutoff distance (option -r). Subunits within the cutoff distance will be explicitly simulated.
A standard run would involve the following flags
-in:file:s
-in:file:native
-database
-symmetry:symmetry_definition
-packing:ex1
-packing:ex2aro
-out:nstruct
-out:file:fullatom
-symmetry:initialize_rigid_body_dofs
-symmetry:symmetric_rmsdIf you want to cluster the result make sure to output silent files ( -out:file:silent and -out:file:silent_struct_type binary ) as the symmetry information is lost when outputting PDB files.
The protocol produces PDB files or a silent file depending on your options together with a scorefile. The normal exit status of the protocol is coming from the jd2 job distributor. It says something like "You'll see the jd2 x jobs completed in y seconds message if successfully completed".
The same procedures used for regular docking are used for symmetric docking. If you have a native reference generate a score vs rmsd plot and look for a funnel. Without a reference structure cluster the 200-400 lowest energy models using the cluster app. For some cases you may want to calculate a symmetric rms during clutering. Add -symmetry:symmetric_rmsd as an option to the cluster app.