1. Opening the protein pdb file.
Go to
File/Read Molecule.
Read in a pdb
file for a protein that has no hydrogen atoms.
Before
you read in the pdb file open it and change the iron atom from FE to Fe.
In general, all element must have the same
appearance as in the periodic table for use in
Autodock.
The protein
will look as below.
2. Assign polar hydrogens and save the
file as a pdbqt file.
Go to
Edit/Hydrogens/Add.
Select “Polar
Only”.
This means O and N atoms which is polar will get
an H.
C-H which is non polar will be treated as a unit in
the electrostatic calculation.
Now we need
to write the file type for the Autodock calculation.
It has a pdbqt extension.
Go to
Grid/Macromolecule/Choose.
Make sure the
molecule is highlighted and click on “Select Molecule”
Write the
file as a pdbqt on the menu that appears.
3. Determine a grid center and grid
dimensions for the Autodock calculation.
Edit the
pdbqt file using an editor like UltraEdit.
Here we want to identify some location in the protein that will serve as
the center of the grid.
I will choose
the Fe atom since I am most interested in the heme.
You could also choose the center of mass of
the protein.
Recall that you can find
the center of mass in VMD using the commands
Ø
set
OBJECT [atomselect top prot]
Ø
measure
center $OBJECT
Next we
choose Grid/Grid Box
The Grid menu
appears.
Adjust the
Spacing to 1.00 so the grid is in even increments of 1.00 Angstrom.
Input the
center of the grid and then adjust the grid size.
The grid dimensions can range from 1 to 40
Angstroms.
The 18 x 18 x
18 grid is appropriate to focus on the heme itself.
The 32 x 40 x
30 grid covers essential the whole protein.
4. Read in the substrate from a pdb
file and then generate the pdbqt file.
Next we need
to read in the substrate.
First, we want
to check the substrate file to make sure that the atom types have the
appropriate names.
For example, bromine
needs to be Fe and not FE.
To load the
substrate into Autodock Tools use the Ligand/Input/Open menu selection.
‘
‘Hide the
protein and center the substrate (just for appearance).
When the
ligand is read into the program, it automatically is assigned charges.
To write the
ligand as an output file use Ligand/Output and then select to write it as a
pdbqt file.
Aside from
writing the pdbqt files, the Autodock Tools interface does not generate the
configuration file for “vina” the docking routine.
5. Edit the configuration file to
prepare to run Vina.
Along the way
you need to save the information you have obtained, such as the center of mass
and grid box and write that information in a very simple configuration file
shown below.
The output
will be put in all.pdbqt as a series of 10 different structures of the
substrate.
The
exhaustiveness tells the program how hard to search.
In our case the molecule is very simple and
the protein is relatively small so we set the exhaustiveness to the maximum value
of 16.
6. Dock the substrate using Vina.
To run vina,
go to the Command Prompt.
Navigate to
the appropriate directory.
One clever
trick is to put the vina program (which you download from the Autodock website)
into the Windows/system32 directory.
From the directory where the program is you type.
Ø
copy
vina.exe c:\windows\system32
Then you may
type vina anywhere on your system and it will run the vina program. If you type
Ø
vina
You get a
kind of help menu.
To run the
program using the configuration file config.txt you type.
Ø
vina
–config config.txt –log log.txt
Note use two
dashes - -, not one. Once vina starts
running you will see the progress on the screen.
It typically takes only a few minutes for our
system.
You can read in the final pdbqt structure using the Read/Molecule command and it will show you all of your docked structures.
The green substrate remains from previous work.
We could also delete it if we want to put the protein in the center of the screen.
An
important final step is to read your docked inhibitor molecules into
VMD and check that the molecular bonds are all in order.
Sometimes
atoms get switched or minus signs get deleted during translation of file
types between pdbqt and pdb.
A common error that we have observed is
for the CAG and CAH carbon atoms to be reversed in the docked inhibitor.
Then the bonds look like a zig-zag across the phenyl ring.
You may
fix this quick simply by editing the pdb file and reversing the CAG and
CAH atom names.
