Docking Box Size

Selecting an appropriate search space is critical to achieve high prediction accuracy in structure-based virtual screening. Therefore, we developed a procedure to customize the box size for individual query ligands in order to maximize the accuracy of molecular docking. Large-scale benchmarking calculations using AutoDock Vina the DUD-E dataset show that using the optimized box size improves the ranking accuracy in virtual screening. These results can help fully automate large-scale virtual screening calculations by customizing docking protocols on the fly for individual library compounds.
If you find our method useful, please cite the following paper:


Source codes

Version (hover) Download Size MD5  

1.1 latest 4.1K 16f502a2756e721e770e6ca2569f8a13  
1.0 4.1K f40d1b434e760e91ab104156d6f86179  



The Perl script accepts ligands in PDBQT, SDF and MOL2 formats and returns a single number that is the optimal edge length of a cubic docking box.

[example]$ nadp.pdbqt


Note that we recently re-optimized eBoxSize on a much larger dataset of pharmacologically relevant protein-drug complexes. The latest version (1.1) gives somewhat larger boxes yielding even better docking results.


Below, we show how to use eBoxSize with Vina by self-docking NADP to aldose reductase (PDB-ID: 1adsA). The required input files are nadp.pdbqt (NADP) and aldr1.pdbqt (aldose reductase). Both files are in the PDBQT format. It is assumed that and Vina are available from the search path.

[example]$ BOX_SIZE=$( nadp.pdbqt)

[example]$ vina --receptor aldr1.pdbqt --ligand nadp.pdbqt --center_x 6.896 --center_y 0.784 --center_z 7.839 --size_x $BOX_SIZE --size_y $BOX_SIZE --size_z $BOX_SIZE --out aldr1-nadp.pdbqt --num_modes 1

PDB-bench dataset
Dataset Download Size MD5  

Target proteins PDB-bench.lst 22k 1f2e8decf7f4e73b14012f2edd2aeb30  

Target structures, ligand binding site predictions and other files for the PDB-bench and DUD-E datasets are available from the eFindSite datasets page.


© Michal Brylinski
This website is hosted at the CCT