The LabelHash Server

ViewMatch

Whether you use the online server or the command line version of LabelHash, you will probably want to visually inspect your matches. We have developed a plugin for Chimera called ViewMatch that allows you to do just that. Below you will find detailed instructions on installing and using this plugin.

Installation instructions for ViewMatch

  • Download and install a recent version of Chimera.
  • You need a local copy of the Protein Data Bank (or at least the PDB files in your match files). See the instructions at the PDB on creating a mirror of the PDB. ViewMatch uses the Chimera Preferences to find PDB files. It will check the path specified as the download directory for fetching (in Favorites>Preferences>Fetch). It will also check the personal PDB directories (specified in Favorites>Preferences>PDB). If ViewMatch still can't find the PDB file for a match, it will attempt to download it. This last option should really be considered a last resort safety net, since it so much slower.
  • To make Chimera detect the ViewMatch plugin, go to the Favorites menu and select Preferences. Select the Tools category in the top left of the Preferences window. Click the Add button and select the directory that contains the ViewMatch directory (not the ViewMatch directory itself). If you plan to use the ViewMatch plugin often, you may wish to add ViewMatch to the Toolbar and your Favorites as shown below. Restart Chimera.
    PrefsTools.png

Using the ViewMatch plugin

The ViewMatch plugin is derived from the ViewDock plugin that comes with Chimera. To load an XML file with matches, go to the Tools menu, select the Structure Comparison submenu, and select ViewMatch. If you have added ViewMatch to the toolbar you can just click on the ViewMatch icon. In the file dialog window that comes up select your file with matches. You can also choose the number of matches that ViewMatch should read. It gets the atom coordinates for the matches from PDB files. After loading is finished, the main Chimera window will show the motif in transparent white superimposed with a match:

ViewMatch.png

The carbon atoms are white in the motif and green in the match. Other atoms in the motif and match are colored by element. The C-alphas are shown as spheres, while the remaining motif and match atoms are shown in stick representation. The rest of the matching protein is shown in ribbon representation. The ViewMatch controller window should look something like this:

ViewMatchController.png

The name of a match is its PDB ID followed by the chain ID. For each match we show the RMSD to the motif, the p-value, the depth, and the score. The depth is the average distance of the match residues to the molecular surface. The score is currently just the number of matching residues. In the web version of LabelHash only complete matches are computed, so this number will be the same for all matches.

The bottom half of the window shows additional information for the currently selected match. This includes the EC class and Gene Ontology (GO) annotations, if known. The PDB-to-EC mapping comes from PDBSProtEC site. The GO terms are taken from the Jena Library. Also included are the REMARK records from the PDB files. Finally, the matching parameters are added. The matching parameters are the same for all matches.

For each match only the matching residues are shown. By pressing the “Load PDB file” button, the entire PDB file is loaded for the selected match and aligned with the motif. By default it shown in the ribbon representation. If more information about certain matches is needed, press the PDBsum button. This will open the corresponding PDBsum web pages in your browser for the selected matches.