Hydrated Docking with AutoDock

The package contains scripts, data files and example structures to perform an AutoDock hydrated docking as described in the paper “A Force Field with Discrete Displaceable Waters and Desolvation Entropy for Hydrated Ligand Docking”, J.Med.Chem. 2012

This implementation requires a certain degree of familiarity with the standard AutoDock protocol and with the command-line interface.

Be aware that with this implementation of the method, it is difficult to compare results obtained
with very diverse ligands without doing extra of post-processing on the results, because the
energy estimation needs to be normalized. For this reason, the method is not suitable for virtual
screenings. This doesn’t affect the structural accuracy, so comparisons within docking poses are fine.
An improved scoring function to overcome this issue is in the works.

For questions and help, please use the AutoDock mailing list.

Happy Dockings!

Download Files

AutoDock_Hydrated_Docking.tar.gz  (4 MB) platform_all_platforms.gif

HYDRATED DOCKING PROTOCOL

– This protocol assumes all the files have been already prepared for a standard docking
protocol (i.e. ligand and receptor structures; GPF and DPF parameter files).

– Help for all the scripts can be accessed by running them with no options.

– The directory ‘example’ contains a case study with both input and output files from a
typical hydrated docking calculation.

1. Add W atoms to the ligands
——————————
The W atoms must be added to a PDBQT file. By default the hydrated ligand is saved with
the “_HYDRO” suffix added (i.e. ligand.pdbqt => ligand_HYDRO.pdbqt).

$ python wet.py -i ligand.pdbqt

2. Calculate grid maps
———————-
Calculate the grid maps following the standard AutoDock protocol, checking that OA and HD
types are present in the ligand atom set. If not, the GPF file must be modified to include
them; i.e. :
– add OA and HD to the line “ligand_types …. ”
– add lines “map protein.HD.map” and “map protein.OA.map”

3. Generate the W map
———————
Water maps are generated by combining OA and HD maps. If standard filenames are used for
maps (i.e. receptor = protein.pdbqt >> maps = protein.OA.map, protein.HD.map), only the
receptor name must be specified:

$ python mapwater.py -r protein.pdbqt -s protein.W.map

4. Run dockings
—————
Prepare the DPF containing the keyword “parameter_file AD4_water_forcefield.dat”, and add
the W type map (“protein.W.map”), then run the docking.

5. Extract and score the results
——————————–
Docking results are filtered by using the receptor to remove displaced waters and the W
map file to rank the conserved ones. By default, the LELC pose is extracted as result.

$ python dry.py -c -r protein.pdbqt -m protein.W.map -i ligand_HYDRO_protein.dlg

Waters are ranked (STRONG, WEAK) and scored inside the output file (“*_LELC_DRY_SCORED.pdbqt”) with the
calculated energy.

REMARK  STRONG water ( score: -0.91 )