DockIt installation instructions

gunzip and untar the dockit_1.5_ARCH.tar.gz file in some convenient directory (where ARCH is either SGI, Sun or linux). This will create a dockit subdirectory and some subdirectories under that. The only other setup you will need to do to run the programs is to define the environment variable M4X_DOCKIT_ROOT to be the full path to the newly created dockit directory. You will also need to place the DockIt license file you were provided into $M4X_DOCKIT_ROOT/etc/license.dat.

cd to dockit/test and run the following commands to make sure everything is working:

../bin/setup 7dfr.pdb

This command will process the 7dfr protein and ligand and do the atom typing for the following docking. You will get a number of warning messages about missing atom types for the cofactor (residue NAD). These can be ignored. They stem from the fact cofactor atom types are not well defined for the PMF scoring function and there is currently no automatic way to define them. The built in docking score function and the PLP score function do assign types to the cofactor.

../bin/run_dock

which will dock the ligand into 7dfr and print out a short report of the results. You will need a terminal window of at least 128 characters wide to see the report in the best format.

The two script files, set and run_dock show how to run DockIt starting from a pdb file which includes a ligand. It is best to start with the protein file including a ligand if possible since that gives the programs the best chance to automatically select the binding site and any necessary waters. However, it is possible to use a protein structure without a ligand as well. In that case, the arguments to sublime would need to be slightly changed in the setup script.

It is possible to run multiple ligands through the docking program. To try this, do the following:

../bin/setup_ligands

which reads in a few ligands from ligands.tdt and creates a CEX file with the appropriate atom type information for docking. In this release, it is easiest to start from a collection of molecules in a Daylight TDT file. Our conversion program reads in the ISM records if they are present and preserves the stereochemistry. Otherwise, generic SMI smiles records are used. If 3D coordinates are present (in $D3D records) then the stereochemistry implied by those coordinates is used. Also, any identifier in a NAM record is passed along. If no stereochemistry information is present, then the docking will randomly sample possible stereochemistries. The example file, ligands.tdt, does not contain ISM records so the stereochemistry of the resulting dock runs is random. To dock the ligands, do:

../bin/run_dock_ligands

which will generate 100 dockings for each ligand in the input. The output is in CEX format, which can be converted back to tdt with the program cex2tdt (in the bin directory) or to pdb format with the included cex2pdb program.

We have also included a version of rasmol which can read CEX format. To view the output from the 7dfr docking, for example, use the command

../bin/rasmol -cex docked.cex

This will read in all 100 docking results as well as the reference (Xray) conformation (as the zeroth model). The rasmol script docklig.spt (in scripts) illustrates how to view a few of the results, compared to the reference. In rasmol use the source command to execute the docklig.spt script. Also check out the utils subdirectory and the documentation for some useful programs to filter out the best dockings from the DockIt output.

By looking at the scripts above you should be able to get a pretty good idea of how to run the programs. Consult the rest of the documentation for further details. Also, our presentation at MUG2000 (http://www.daylight.com/meetings/mug2000/Dixon/dockit.html) might help.