Code for this application is in src/apps/public/design/zinc_heterodimer_design (setup and driver) and src/protocols/metal_interface/ZincHeterodimerMover (mover). The integration test is called zinc_heterodimer.
[http://www.ncbi.nlm.nih.gov/pubmed/23504819] Der BS, Jha RK, Lewis SM, Thompson PM, Guntas G, Kuhlman B. Combined computational design of a zinc-binding site and a protein-protein interaction: one open zinc coordination site was not a robust hotspot for de novo ubiquitin binding. Proteins. 2013 Jul;81(7):1245-55. doi: 10.1002/prot.24280. Epub 2013 Apr 20. Erratum in: Proteins. 2013 Sep;81(9):1678. Jha, Raamesh K [corrected to Jha, Ramesh K]. PubMed PMID: 23504819; PubMed Central PMCID: PMC4084500.
This protocol designs zinc-mediated heterodimers. One partner contains a three-residue zinc binding site, the second partner contains one surface zinc-coordinating residue. In the example given, partner 2 is ubiquitin, which contains a surface histidine.
Prior to running this protocol, RosettaMatch is used to design three-residue zinc sites on a protein surface. Next, the target protein containing a surface histidine (Ubq-H68) is docked to the open coordination site of zinc. The ubiquitin rigid-body orientation to the designed zinc-binding scaffold is sampled as if the ubiquitin were a giant protein-sized rotamer. This is done using inverse-rotamer sampling of Ubq-H68 chi1 and chi2 angles (SidechainMover), as well as free rotation about the His68-zinc coordination axis (RotateJumpAxisMover). There is a centroid phase that evaluates overall shape complementarity of the target and scaffold, this phase is coupled to a full-atom pose to remember the conformation of the zinc coordinating residues.
The lowest-scoring centroid pose is the only pose that enters into full-atom interface design. Perhaps a better way to do it would be to design every non-clashing backbone orientation. There is no backbone minimization or sampling in this protocol. The zinc site remains fixed during interface design.
The biggest difference in 'mode' is whether or not you want to output all of the rigid-body perturbed structures for viewing/debugging. See command-line options.
These values are for production runs.
-resfile resfile #the resfile default is NATAA, this will prevent mutation of the wild-type target. For the scaffold, all residues should be NOTAA HC.
-skip_sitegraft_repack false #the local region surrounding the zinc match residues is repacked by default. #-AnchoredDesign::perturb_temp
-AnchoredDesign::perturb_cycles 500 #how many times to perturb the rigid-body orientation
-AnchoredDesign::perturb_show false #true for debugging/visualization of rigid-body sampling
-AnchoredDesign::refine_cycles 10 #how many times to run PackRotamersMover
The expected output is a heterocomplex containing: target protein, zinc, zinc binding site, designed scaffold. If the pertub_show option is used, you'll see PDBs from each rigid-body sampling step. The log file will show you how many times SidechainMover or RotateJumpAxisMover were called, and whether the MonteCarlo move was accepted or rejected. The end of the log file prints a table of energies.
After running the protocol, the best way to evaluate designs is by computed binding energy using InterfaceAnalyzer (specify a jumpnum of 2). Take a look at the zinc site to make sure that 4-residue coordination (3-by-1) is intact.
If you've made improvements, note them here.