The standalone version of DDB can be downloaded in the form of Python command line programs and a PyMOL plugin.

 

  1. Python programs (Command Line Interface).

    • Installation instructions
    • Screenshots of DDB CLI
  2. ddb_cli_ubuntuScreenshot

  3. PyMOL Plugin (Graphical User Interface).

    • Installation instructions
      • This Plugin works on Linux Ubuntu 12.04 (x64) or later. We will release a version that can work on Windows and OSX in the near future.
      • To install the plugin automatically via PyMOL’s Plugin Manager, launch PyMOL. Go to the main menu > Plugin > Plugin Manager >  Install New Plugin > Install from local file > Choose file …  and then browse to ddb.zip you downloaded from here.
      • You will be asked to install the plugin somewhere in your home directory, we recommend accepting the suggested home-directory-based path (e.g., ~/.pymol/startup/) since you will be able to write files to that location without permission restrictions.
      • To install the plugin manually, please find the path of PyMOL plugins and unzip the plugin file ddb.zip  in there. In some Ubuntu Linux systems, this path is /usr/lib/python2.7/dist-packages/pmg_tk/startup/. We however recommend that you install it in the home directory (such as ~/.pymol/startup) where you have permissions to write files.
      • The three descriptor extraction tools NNSCORE, SMINA, and AUTODOCK41 require the protein and ligand files converted into PDBQT format. We will use MGLTools from scripps institute to convert these molecules from PDB & MOL2 formats to PDBQT. A compressed version of the software is included in the utils/ subdirectory. Please follow the instructions to install the software:
        • $cd /path/to/ddb/utils/
          • Remember to replace /path/to/ddb/utils with the actual ddb path such as ~/.pymol/startup/ddb/utils/
        • $mkdir mgltools
        • $tar -xvf mgltools_x86_64Linux2_1.5.6.tar.gz
        • $cd tar -xvf mgltools_x86_64Linux2_1.5.6.tar.gz
        • $./install -d /path/to/ddb/utils/mgltools
          • Remember to replace /path/to/ddb/utils with the actual ddb path such as ~/.pymol/startup/ddb/utils/
        • Change the first line of the files /path/to/ddb/utils/mgltools/MGLToolsPckgs/AutoDockTools/Utilities24/prepare_ligand4.py and and /path/to/ddb/utils/mgltools/MGLToolsPckgs/AutoDockTools/Utilities24/prepare_receptor4.py from #!/usr/bin/env python to #!/usr/bin/env pythonsh
    • Sample datasets
      • Please note that in the plugin directory /path/to/ddb, we include a data directory that contains pre-extracted set of descriptors for the primary training set and the core test set complexes of PDBbind 2014. You can employ these data sets to train and evaluate machine learning scoring functions using the Fit & Predict utility of the DDB plugin. The paths to these data sets are: /path/to/ddb/data/descriptors/ScorePDBbind14PrTr/ and /path/to/ddb/data/descriptors/ScorePDBbind14CrTs/. The names and activities values of these complexes are stored in the files “molNames_and_responseValues.csv” located inside these directories.
      • We also include a sample of ten protein-ligand complexes from PDBbind to test the descriptor extraction utillity of DDB. These molecules can be found in /path/to/ddb/data/plc_data/pdbbind15s.
    • Screenshots of DDB Plugin

      DDB_pymol_DE

      DDB_pymol_filtering
      DDB_pymol_fitPredict

      DDB_pymol_About