Virtual screening with Autodock-4, Autodock-Vina, eHiTS, GOLD or DOCK. Analyzing with Dockres.

M. Mezei & M.-M. Zhou, Source Code for Biology and Medicine, 5, 2 (2010).     Get copy

Last updated: 01/19/2012

Licensing/Downloading Information	Library screening documentation	Download distribution
	Dockres documentation

Major features implemented since the publication of the paper:

• Support for screening with eHiTS
• Option to run Dockres in batch mode using command-line options (5/2010)
• Parallelized version of Dockres (6/2010)
• Support for analyzing the results of screening with GOLD (9/2010)
• Support for virtual screening and analysis using Autodock-Vina, (12/2010)
• Option to submit multiple screenings with running the script once (09/2011)
• Option to filter posed by overlap to a target structure (10/2011)
• Support for further filtering of top-scoring DOCK ligands(01/2012).

NOTE: New features are still undergoing testing - please report any problem you may have encountered.
NOTE: Previous versions are saved on the website - in case of problem they can be tried.

LIBRARY SCREENING: A collection of utilities are in the subdirectory "fullscreen" of the distribution directory. The most important one, fullscreen.csh performs the screening. It takes as input

• A macromolecule/receptor file in Autodock's .pdbqs or .pdbqt format or in Tripos' .mol2 format
• For all Autodock versions a directory name containing the ligand files in either Tripos' .mol2 format or in Autodock's .pdbqt format
• For eHiTS a single file containing all igand descriptions in any of .pdb,.,ol2, or .sdf formats.
• Information about the docking region/grid
• Various parameters needed for the run, including the number of CPU's.

NOTE, that several of the scripts in this collection require customization to specify the location of various directories.

POST-PROCESSING: The mainstay of post-processing is the program Dockres. It reads the log files created by Autodock (version 3.0.5 and version 4) or the pdbqt files created by Autodock-Vina or the .sdf files created by eHiTS or the .mol2 files created by GOLD and extracts the top scoring poses.

Dockres does not require that the screening is done with fullscreen.csh - it runs on any set of log files.

The extraction can be subject to various filters, either based on the receptor (e.g., the residue nearest to the ligand, the closest approach to a preselected macromolecule atom) or on the ligand (e.g., charge, molecular weight). These filters can also be applied to the top-scoring ligands generated by DOCK.

The program also calculates distributions of various properties of the ligand set (e.g., molecular weight, number of hydrogen bond donors) and the distribution of docking sites as well as the distribution of docking free energies per target residue.

Also, selected poses can be extracted into PDB files that contains the macromolecule as well.

Go to the Mezei Lab home page