Yes, AutoDock can be used when the location of the binding site is unknown. This is often referred to as "blind docking", when all that is known is the structure of the ligand and the macromolecule.

It will be necessary to set up the dockings to search the entire surface of the protein (or other macromolecule) of interest. This can be achieved using AutoGrid to create very large grid maps, with the maximum number of points in each dimension, and if necessary, creating sets of adjacent grid map volumes that cover the macromolecule. The third-party tool BDT can be used to set up such sets of grid maps.

Several authors have used AutoDock to perform blind docking (see 1.-6.); for instance, Hetenyi et al. published two papers showing that AutoDock can be used to perform blind docking of peptides to proteins, and drug-sized molecules to proteins.

1. Hetenyi, C. and van der Spoel, D. (2002) Efficient docking of peptides to proteins without prior knowledge of the binding site. Protein Science, 11(7): 1729-1737.
2. Kovacs, M., Toth, J., Hetenyi, C., Malnasi-Csizmadia, A., and Sellers, J.R. (2004) Mechanism of blebbistatin inhibition of myosin II. Journal of Biological Chemistry, 279(34): 35557-35563.
3. Bikadi, Z., Hazai, E., Zsila, F., and Lockwood, S.F. (2006) Molecular modeling of non-covalent binding of homochiral (3S,3 ' S)-astaxanthin to matrix metalloproteinase-13 (MMP-13). Bioorganic & Medicinal Chemistry, 14(16): 5451-5458.
4. Hazai, E., Bikadi, Z., Zsila, F., and Lockwood, S.F. (2006) Molecular modeling of the non-covalent binding of the dietary tomato carotenoids lycopene and lycophyll, and selected oxidative metabolites with 5-lipoxygenase. Bioorganic & Medicinal Chemistry, 14(20): 6859-6867.
5. Hetenyi, C. and van der Spoel, D. (2006) Blind docking of drug-sized compounds to proteins with up to a thousand residues. FEBS Letters, 580(5): 1447-1450.
6. Iorga, B., Herlem, D., Barre, E., and Guillou, C. (2006) Acetylcholine nicotinic receptors: finding the putative binding site of allosteric modulators using the "blind docking" approach. Journal of Molecular Modeling, 12(3): 366-372.

In AutoDock 4, we have introduced a new command "parameter_file" that takes a new parameter library file that contains the various force field parameters.  In most cases, you will not need to modify these values, but it is important to know where they are and how to change them if necessary.

The standard AutoDock 4 parameters are in the file "AD4_parameters.dat" which can be found in the "autodocksuite-4.n.m/src/autodock-4.x.y" directory of the AutoDock 4 distribution, where n and m are the major and minor version numbers of the AutoDock Suite, and x and y are the major and minor version numbers of AutoDock.  Here is an example:

# $Id: AD4_parameters.dat,v 1.14 2007/04/27 06:01:47 garrett Exp $
# 
# AutoDock 
# 
# Copyright (C) 1989-2007,  Garrett M. Morris, David S. Goodsell, Ruth Huey, Arthur J. Olson, 
# All Rights Reserved.
# 
# AutoDock is a Trade Mark of The Scripps Research Institute.
# 
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License
# as published by the Free Software Foundation; either version 2
# of the License, or (at your option) any later version.
# 
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU General Public License for more details.
# 
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software
# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.

# AutoDock Linear Free Energy Model Coefficients and Energetic Parameters
#                        Version 1.0
#                     $Revision: 1.14 $

# AutoDock 4 free energy coefficients with respect to original (AD2) energetic parameters
#
#               Free Energy Coefficient
#               ------
FE_coeff_vdW    0.1560
FE_coeff_hbond  0.0974
FE_coeff_estat  0.1465
FE_coeff_desolv 0.1159
FE_coeff_tors   0.2744

# AutoDock 4 Energy Parameters

# - Atomic solvation volumes and parameters
# - Unweighted vdW and Unweighted H-bond Well Depths
#
# - Atom Types
# - Rii = sum of vdW radii of two like atoms (in Angstrom)
# - epsii = vdW well depth (in Kcal/mol)
# - vol = atomic solvation volume (in Angstrom^3)
# - solpar = atomic solvation parameter
# - Rij_hb = H-bond radius of the heteroatom in contact with a hydrogen (in Angstrom)
# - epsij_hb = well depth of H-bond (in Kcal/mol)
# - hbond = integer indicating type of H-bonding atom (0=no H-bond)
# - rec_index = initialised to -1, but later on holds count of how many of this atom type are in receptor
# - map_index = initialised to -1, but later on holds the index of the AutoGrid map
# - bond_index = used in AutoDock to detect bonds; see "mdist.h", enum {C,N,O,H,XX,P,S}
#
# - To obtain the Rij value for non H-bonding atoms, calculate the 
#        arithmetic mean of the Rii values for the two atom types.
#        Rij = (Rii + Rjj) / 2
#
# - To obtain the epsij value for non H-bonding atoms, calculate the 
#        geometric mean of the epsii values for the two atom types.
#        epsij = sqrt( epsii * epsjj )
#
# - Note that the Rij_hb value is non-zero for heteroatoms only, and zero for H atoms;
#        to obtain the length of an H-bond, look up Rij_hb for the heteroatom only;
#        this is combined with the Rii value for H in the receptor, in AutoGrid.
#        For example, the Rij_hb for OA-HD H-bonds will be (1.9 + 1.0) Angstrom, 
#        and the weighted epsij_hb will be 5.0 kcal/mol * FE_coeff_hbond.
#
#        Atom   Rii                             Rij_hb       rec_index
#        Type         epsii           solpar         epsij_hb    map_index
#                            vol                          hbond     bond_index
#        --     ----  -----  -------  --------  ---  ---  -  --  -- --
atom_par H      2.00  0.020   0.0000   0.00051  0.0  0.0  0  -1  -1  3    # Non H-bonding Hydrogen
atom_par HD     2.00  0.020   0.0000   0.00051  0.0  0.0  2  -1  -1  3    # Donor 1 H-bond Hydrogen
atom_par HS     2.00  0.020   0.0000   0.00051  0.0  0.0  1  -1  -1  3    # Donor S Spherical Hydrogen
atom_par C      4.00  0.150  33.5103  -0.00143  0.0  0.0  0  -1  -1  0    # Non H-bonding Aliphatic Carbon
atom_par A      4.00  0.150  33.5103  -0.00052  0.0  0.0  0  -1  -1  0    # Non H-bonding Aromatic Carbon
atom_par N      3.50  0.160  22.4493  -0.00162  0.0  0.0  0  -1  -1  1    # Non H-bonding Nitrogen
atom_par NA     3.50  0.160  22.4493  -0.00162  1.9  5.0  4  -1  -1  1    # Acceptor 1 H-bond Nitrogen
atom_par NS     3.50  0.160  22.4493  -0.00162  1.9  5.0  3  -1  -1  1    # Acceptor S Spherical Nitrogen
atom_par OA     3.20  0.200  17.1573  -0.00251  1.9  5.0  5  -1  -1  2    # Acceptor 2 H-bonds Oxygen
atom_par OS     3.20  0.200  17.1573  -0.00251  1.9  5.0  3  -1  -1  2    # Acceptor S Spherical Oxygen
atom_par F      3.09  0.080  15.4480  -0.00110  0.0  0.0  0  -1  -1  4    # Non H-bonding Fluorine
atom_par Mg     1.30  0.875   1.5600  -0.00110  0.0  0.0  0  -1  -1  4    # Non H-bonding Magnesium
atom_par MG     1.30  0.875   1.5600  -0.00110  0.0  0.0  0  -1  -1  4    # Non H-bonding Magnesium
atom_par P      4.20  0.200  38.7924  -0.00110  0.0  0.0  0  -1  -1  5    # Non H-bonding Phosphorus
atom_par SA     4.00  0.200  33.5103  -0.00214  2.5  1.0  5  -1  -1  6    # Acceptor 2 H-bonds Sulphur
atom_par S      4.00  0.200  33.5103  -0.00214  0.0  0.0  0  -1  -1  6    # Non H-bonding Sulphur
atom_par Cl     4.09  0.276  35.8235  -0.00110  0.0  0.0  0  -1  -1  4    # Non H-bonding Chlorine
atom_par CL     4.09  0.276  35.8235  -0.00110  0.0  0.0  0  -1  -1  4    # Non H-bonding Chlorine
atom_par Ca     1.98  0.550   2.7700  -0.00110  0.0  0.0  0  -1  -1  4    # Non H-bonding Calcium
atom_par CA     1.98  0.550   2.7700  -0.00110  0.0  0.0  0  -1  -1  4    # Non H-bonding Calcium
atom_par Mn     1.30  0.875   2.1400  -0.00110  0.0  0.0  0  -1  -1  4    # Non H-bonding Manganese
atom_par MN     1.30  0.875   2.1400  -0.00110  0.0  0.0  0  -1  -1  4    # Non H-bonding Manganese
atom_par Fe     1.30  0.010   1.8400  -0.00110  0.0  0.0  0  -1  -1  4    # Non H-bonding Iron
atom_par FE     1.30  0.010   1.8400  -0.00110  0.0  0.0  0  -1  -1  4    # Non H-bonding Iron
atom_par Zn     1.48  0.550   1.7000  -0.00110  0.0  0.0  0  -1  -1  4    # Non H-bonding Zinc
atom_par ZN     1.48  0.550   1.7000  -0.00110  0.0  0.0  0  -1  -1  4    # Non H-bonding Zinc
atom_par Br     4.33  0.389  42.5661  -0.00110  0.0  0.0  0  -1  -1  4    # Non H-bonding Bromine
atom_par BR     4.33  0.389  42.5661  -0.00110  0.0  0.0  0  -1  -1  4    # Non H-bonding Bromine
atom_par I      4.72  0.550  55.0585  -0.00110  0.0  0.0  0  -1  -1  4    # Non H-bonding Iodine

These are the default values that are "baked in" to AutoGrid 4 and AutoDock 4, and are the values that will be used if you do not specify your own parameter file explicitly in your GPF or DPF with the "parameter_file" command.

You can add new atom types and parameters to this file, or modify the existing ones.  If you come up with new or modified parameters, we would like to know about them, so we can incorporate them in future releases.  Thanks!

Yes, for both the macromolecule and the ligand, if you want to use the AutoDock 4 force field properly, you should always add all hydrogens, compute Gasteiger charges and then  merge the non-polar hydrogens.  This is because AutoDock 4 uses the united atom model to represent molecules, and the AD4 scoring function was calibrated using Gasteiger partial charges on both the ligand and the macromolecule.

Polar hydrogens are hydrogen atoms that are bonded to electronegative atoms like oxygen and nitrogen.  (ADT assumes that non-polar hydrogens are hydrogens bonded to carbon atoms.)

Using ADT

There is a command in ADT to help you decide on the protonation of the Histidines, but you have to load the commands before you can use it: go to "File > Load Module" and then scroll down, click on "repairCommands", and then click "Load Module" followed by "Dismiss". Now, go to "Edit > Hydrogens > Edit Histidine Hydrogens".

If there are any histidines in your molecule, a panel will open up listing each histidine residue along with a row of radio buttons. You can use these to choose whether each histidine should be neutral, HD1; neutral, HE2; or protonated.

Using Reduce/Molprobity

There is a very nice tool called Reduce (with a web-accessible front end called Molprobity) that can be used for adding hydrogens and optimising the hydrogen-bond network by flipping amido groups in Asn and Gln sidechains, and His imidazole rings by 180º. It can also be used for evaluating the quality of your protein structure. See:

Word, et al. (1999) "Asparagine and glutamine: using hydrogen atom contacts in the choice of sidechain amide orientation" J. Mol. Biol. 285, 1733-1745.

The various parameters in the AutoDock 4 scoring function are described in another FAQ, Where do I set the AutoDock 4 force field parameters?. The atom types' names and parameters are specified in a file that can be called anything, but by default is called "AD4_parameters.dat". You can find a copy of this default file in the source code of AutoGrid and AutoDock. This parameter file can be specified by the "parameter_file" keyword in the GPF and DPF, but if this keyword is not given, then AutoGrid and AutoDock use the default values. So the only time you need to use the "parameter_file" keyword is when you want to change the default values, or to add new atom types.

How do I add new atom types?

Important: Bear in mind that the AutoDock 4 scoring function was calibrated for the current set of atom types, and that if you add new ones, strictly speaking you should perform a re-calibration of the force field to determine the correct coefficients for the molecular mechanics terms and the empirical term of the linear free energy model.

You will need to find, compute or set the following values for each new atom type, which you specify after the "atom_par" keyword. Note that the values on each "atom_par" line are space delimited, not fixed width. You will need to add one "atom_par" line for every new atom type. It is possible to define 'synonyms' for atom types, by repeating the numerical atom parameter lines for every variant of the atom type's one- or two-character name.

• name of the atom type; this can be one or two characters long, and should correspond to the atom type at the end of ATOM or HETATM lines in PDBQT files; specify this in the "Atom Type" field
• van der Waals radius of the atom/ion (in Angstrom); specify twice this value in the "Rii" field
• epsilon or energy well depth for two like interacting atoms/ions (in Kcal/mol); specify in the "epsii" field
• volume of the atom/ion (in Angstrom^3); compute this from 4/3 * PI * (Rii/2)^3; specify in the "vol" field
• atomic solvation parameter of the atom/ion; the a_i or "ASP" values in the equation for S_i in the description of the Desolvation Free Energy Term in AutoDock 4; specify this in the "solpar" field
• hydrogen bonding radius (in Angstrom); for non-hydrogen-bonding atom types this is 0.0; specify this in the "Rij_hb" field
• hydrogen bonding energy well depth (in Kcal/mol); for non-hydrogen-bonding atom types this is 0.0; specify this in the "epsij_hb" field
• hydrogen bonding type; an integer; for non-hydrogen-bonding atom types this is 0; specify this in the "hbond" field
• receptor type index; an integer; default value is -1; specify this in the "rec_index" field
• grid map index; an integer; default value is -1; specify this in the "map_index" field
• bond type index; an integer; default value is 4; specify this in the "bond_index" field

The last six values (0.0 0.0 0 -1 -1 4) will be the same if the atom type is not involved in hydrogen bonding.

Note that a comment can be specified after a number sign symbol, #

Remember to use the "parameter_file AD4_parameters.dat" command in both the GPF and DPF, so that the new parameters are used in both the AutoGrid pre-calculation of the grid maps, and in the AutoDock dockings. Specify this command on the first line of the GPF and DPF files.