An Enhanced Virtual Screening Approach based on Pocket Ligand Similarity

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 3 of the License, or (at your option) any later version.

EViS is an enhanced virtual screening method which integrates ligand docking, protein pocket template searching and ligand template shape similarity calculating. A novel PL-score to characterize local pocket-ligand template similarity is used to evaluate the screening compounds. The docking poses (MOL2 format) obtained by any docking programs can be evaluated by PL-score comparing with the native pocket-ligand pairs in BioLip dataset.

• Before running EViS, the following packages should be installed.
• gcc: gcc-8.3.0
• python version : 2.7
• numpy: numpy 1.16.6
• git clone: https://github.com/JingHuangLab/EViS.git

• git clone https://github.com/JingHuangLab/EViS.git

• running the following commands

cd EViS 
tar -xvzf user_input.tar.gz 
module load gcc/gcc-8.3.0 

EViS program needs two consecutive steps (please ensure the last step was completed before running the next step):

1, prepare the screening compounds (MOL2 format) and pocket ('.POC' format) into the input folder.

• put all compounds (MOL2 format) into (user_input/results_docking/ligand_mol2/) subfolder.

• put the pocket file (named as "nativepocket.poc") into the (user_input/results_docking/pocket/) subfolder.

  The pocket is like as:

POC     nativepocket
ATOM    253  N   VAL A  49       6.937   3.497 -28.457  1.00 34.68           N
ATOM    254  CA  VAL A  49       6.465   2.179 -28.887  1.00 34.68           C
ATOM    255  C   VAL A  49       5.665   2.297 -30.179  1.00 34.68           C
ATOM    256  O   VAL A  49       5.965   1.651 -31.192  1.00 34.68           O
...
TRE

If you are still not sure the input files, you can check the example files in this repository.

2, runEViS.py

For exmple：

cd script
python run_EViS.py

output:

pocket folder:

• pocket_align.POC : all pocket templates for nativepocket.poc with similarity >0.5
• pocket_topN.info : top 50 (maximum) template pockets information
• pocket_topN.mol2 : the template ligands binding with the topN template pockets in BioLip dataset.

ligands_mol2 folder:

• .lsalign ： the similarity file between the screening compound and topN template ligands save in "pocket_topN.mol2"

results_docking folder:

• PLscore.result : the final PL-score results for the screening compounds.

If you have any questions, please contact with Wenyi Zhang([email protected])

• Wenyi Zhang ([email protected])
• JingHuang ([email protected])

Inspiration, code snippets, etc.

• LS-align
• PPS-align
• ITASSER