DockVision Tour
Hello, welcome to the DockVision guided tour.
We're going to illustrate how to set up and run a docking calculation
with DockVision. The only things you need are:
a ligand coordinate file (PDB format)
a target coordinate file (PDB format)
the X,Y,Z coordinate of an active site atom
Setup
The first step is to start up the program, and then select Research in
the Mode pulldown menu. You are now in the Setup menu for
Research. Type in the run name (this determine the names of the output files)
and select the forcefield.
Your window will look like this:
Ligand
Now, you can proceed to assign the ligand coordinates. To go to the ligand
screen, either select the Ligand box on the left side of the window,
or select the Next button on the lower right. Hit the Load Ligand
button to open a file browser. You can now choose your ligand coordinate
file:
Flexibility
The next screen involves selecting flexibility options for the ligand.
If you select Flexible Ligand, then you can choose to generate
a Topology file (required) and also decide whether or not you want
Conformers and, if so, how many to generate. If you want to dock
the ligand starting from the conformation represented in the coordinate
file (but still allowing the conformation to change during docking) then
selectNo Conformers. If you want to dock starting from random
alternate conformations, then select Generate Automatically
under the conformers menu.
Target
In this screen you select the protein coordinate file to which the
ligand will be docked. Select the Load Target button to activate
the file browser and select your target pdb file.
Grid
In this screen, you select parameters which determine the pre-calculated
grid. Normally, this grid will only need to be calculated once for
each target. If this is the first time you've used this target, select
Auto Generate Grid at the top of the form. By default,
Auto Bounds will be activiated, which will generate a grid covering
the entire protein volume. In general you don't need a grid this large,
because docking is localized around an activite site. Here, you turn
Auto Bounds off and type in the active site X,Y,Z coordinates
with a 15 angstrom size for the grid. Select No Near Grid as this
is no currently used in Research mode.
Options
In this screen, you select specific options which control how Research
behaves. First, you need to build or load a constraint set. This is
a set of spheres that determine where Research will search for binding
modes. The geometrical center of the ligand will be restricted to
remain inside these spherical regions. Next, you select an annealing
schedule. You can either load a previously defined schedule or make a
new one. These determine the temperature, number of steps and maximum
translation and rotation for the Monte Carlo procedure.
There are two important options which need to be set before docking
can proceed. The first is ntrials. This determines the total
number of independent Monte Carlo docking trials to run, and will
depend on the size of the ligand, the size of the search region and
the flexibility of the ligand. The second option is ecut. This
determine the energy below which docked ligands will be saved to the
output file. Since you don't often know in advance how low the energies
may be (this will also depend on whether Conformational Energy
is selected), it is often useful to first perform a short run with
ntrials set to 100 and and ecut set to something large,
like 1000.0. Then look at the energy distrbution using the
Energy Histogram in the Tools menu. Normally, you want
to set ecut so you save about 5% of all trials.
Research
Now you're ready to start running Research. Select the Research
button or Next in the Options. Now select Run.
The run begins by running some automated setup programs, which may
include chargeit, otto, confman and flgrid.
Normally, these programs take only a few seconds to run, but flgrid
may take a few minutes, depending on the size of the grid to be generated.
The progress bar at the bottom left corner of the window will tell
you the percentage of the computation completed.
Once the setup programs are completed, Research itself begins to run.
The output in the main window gives a verbose description of which
stage it is in the startup. Once it reports "Starting Monte Carlo runs..."
the generation of sucessive Monte Carlo trials has begun. Again the
progress bar will tell percentage of trials completed.
If you discover you made an error in selecting any of the parameters,
you can hit Kill to stop the job. You can then go back and
correct any parameter assignments, and then come back and start it from
the beginning again.
JobControl
By selecting different run names, you can simultaneously run a number of
docking simulations. You can track the progress of different processes
by selecting the JobControl item in the Mode pulldown menu
at the top of the window. The status of each process can be checked
by selecting the appropriate process ID in the left window. The selected
process can be killed by clicking on the Kill button, or run
at a lower priority by clicking on the Renice button. Sucessively
selecting Renice does not lower the priority any further.
Energy Histogram
Another way you can track the progress of your run is by using the
Energy Histogram utility in the Tools menu. Go to
the Mode pulldown menu and select Tools.
First select the output file you want to monitor. Select the
Input File button on the top of the menu to pick the file. Output
coordinate files are named "jobname.out". Now select Redraw
to display the energy distribution for your output file. The X axis
represents energy in kcal/mole, while the Y axis represents the number
of dockings which appear in each energy range. You can rescale the
plot by selecting alternate energy ranges and number of bins.
The dockings on the left of the plot represent the lower energy and hence
more favorable dockings. In this plot you can see there are several
good low-energy dockings.
Statistics
You can get a summary of the statistics of the docking run by selecting
Statistics in the Tools menu. Again, just use the file
browser to select the appropriate output file.
Gamma
Running the Gamma docking program is very similar to running
Research. One difference is in the Options menu.
To run Gamma, select Gamma in the Mode pulldown, and
go through the steps of selecting a runname, ligand, etc as you do
with Research. Once you get to the Options menu, things change
because you now control the parameters specific to the Genetic
Algorithm method. Options actually contains three submenus,
Main Options, GA Options and Energy Options.
However, you can use the defaults for the last two submenus and only
be concerned with the Main Options submenu.
The Constraint menu is used just as in Research to
define the search region. Now select Genop File to load in
a set of genetic operators. You can find a file in the "Parms"
directory, which may be accessed in the file browser. (Examine
this file in detail if you want to make your own).
Now select the Number of Generations. This defines how many
sucessive replication and pruning cycles will be executed. Next,
select the Max Population Size and Min Population Size.
The difference between these two represent the number of new
"organisms" generated at each cycle. A good rule of thumb for this
is the difference should be about 10% of the max population size:
hence, for a max population size of 100, you should use a min
population size of 90.
You may wish to use the Subpopulations to specify several
"semi-independent" populations of organisms which only infrequently
interact. This can be used to maintain diversity in order to avoid
the whole population prematurely converging to a local energy
minimum.
Here's the Gamma menu. After Run has been clicked, the
progress of each generation is reported on the screen. Here the
generation number is given, along with the best, median and average
fitnesses. The fitness is simply the negative of the energy.
GA View
The progress of the Gamma run may be monitored by selecting the
GA View utility in the Tools menu. Select a ".graph"
file for the run (again, the file will be named "jobname.graph")
and click on Redraw. The X axis represents the generation number
while the Y axis represents the fitness on a reverse scale, so larger
fitness appear lower in the graph. Large drops in the graph indicate
shifts to new, lower energy docking modes.
Final Comments
Thank you for taking this brief DockVision guided tour! If you have
any questions about obtaining or using this program, please feel
free to contact us at:
support@dockvision.com.
Thanks and happy docking!
Last modified: April 13, 2000