The aim of this tutorial to test the ability of protein-protein docking servers to reproduce experimental complex conformations.

Redocking:

• Select from the PBD a few (say, three) complexes
• Use Simulaid to pick up the biological dimers
• Separate the complex PDB file into two individual PDB files
• Submit each pair to the servers
• Analyze the results.

Unbound docking:

• Select a few (say, three) complexes from the decoy sets
• Submit the individual structures to the servers
• Analyze the results.

Analysis:

• Compare the top scoring result with the experimental complex structure
• If the RMSD between the top-scoring and experimental complexes is more than 10 Å then see if there is any structure among the complexes generated that is within 10 Å
• Think of ways to chracterize the interfaces
• There is a program called rescore that (among others) calculates the RMSD between the experimental structure and the models from ClusPro and PatchDock; the documentation has some pointers how to extract the results.
  It can be execured on Minerva as ~mezeim01/simulaid/rescore.

Docking servers:

• ClusPro: https://cluspro.bu.edu/login.php
• PatchDock: https://bioinfo3d.cs.tau.ac.il/PatchDock/
• Gram-X: http://vakser.compbio.ku.edu/resources/gramm/grammx/

• DOCKGROUND: http://dockground.bioinformatics.ku.edu/UNBOUND/decoy/decoy.php
• ZLAB: https://zlab.umassmed.edu/zdock/decoys.shtml

Last modified: 11/30/2017          Go to Mezei Lab Home page