Welcome to the Gray Lab
About Us Research Publications Lab Members Facilities Positions Links
Docking Decoys
There are three sets of decoys:
  1. "bound perturbations": from perturbations of the bound complex structure (with native sidechains removed and rebuilt from rotamers). 1000 decoys for each of 54 targets. bound_perturb.tgz (3.3G)
  2. "unbound perturbations": from perturbations of the unbound component structures which were superimposed on the bound backbones (again, with sidechains removed and rebuilt). 1000 decoys for each of 54 targets. unbound_perturb.tgz (3.3G)
  3. "global": top 200 structures from global searches starting from unbound starting structures with rebuilt sidechains. These sets may or may not have near-native decoys. 200 decoys from each of 54 targets. unbound_global.tgz (677M)
Reference: J.J.Gray et al. J.Mol.Biol. 331(1), 281-299, 2003 "Protein-Protein Docking with Simultaneous Optimization of Rigid-body Displacement and Side-chain Conformations" PDF

Docking benchmark set from http://zlab.bu.edu/~rong/dock/benchmark.shtml
Other notes:
  • docking partners are separated by TER lines. Multiple chains in the same docking partner (e.g. heavy and light chains of an antibody) will NOT have a TER line between them.
  • rms and score values are tabulated in the score files (*.fasc) in columns 3 and 2, these should match the points shown in Figs 3-5 of the JMB paper.
  • natives have been taken from Chen's benchmark set, stripped of any residues with missing backbone atoms, and combined with the TER convention described above.
Some oddities (Thanks to Yaoqi Zhou for noticing):
JJG 07/06/03, rev. 08/18/03   graylab.jhu.edu  bakerlab.org