This is Penn State

Supplementary material for DockRank ACM-BCB 2011 paper

 
The scatter plot of rankings for the docking cases that can be predicted by PS-HomPPI.pdf

Please note that ACM-BCB 2011 paper used docked models generated by Cluspro 2.0 using Electrostatic-favored energy functions.

Cluspro 2.0 - BM3 decoy set:
  • The PDB files of docked models
  • The mapping between Cluspro task IDs and case IDs
  • The L-RMSD of docked models
  • Cluspro Energy Scores
  • DockRank Scores
  • Some notes about this decoy set:

    1. Two of the docked models returned by ClusPro 2.0 are abnormal. Several lines of PDB files are combined into one line.

          Dataset: Docking Benchmark 3.0

          Case ID: 1KTZ

          model ID: model.000.10

          ClusPro job ID: 6977

          Dataset: Docking Benchmark 3.0

          Case ID: 1CGI

          model ID: model.002.10

          ClusPro job ID: 6902

    2. We found for two cases, 1RLB and 1ML0, one of the two ligands were left out by the authors of the BM3 dataset. Each of the case should have two identical bound ligands, however, the original BM3 set only include the coordinates of one ligand as bound ligand, which may cause docked models near the other bound ligand position regarded as wrong models. Here we attach the bound PDB files for case 1RLB and 1ML0. We also recalculated the L-RMSD for all the docked models by including both of the two identical bound ligands. The smaller value of L-RMSDs between a docked ligand and two bound ligands is used as final L-RMSD for this model.

  • 1RLB
  • 1ML0
  •