Installation (Requires at least Python version 3.9 )

Use the following command in terminal (Mac/Linux):

git clone https://github.com/NRGlab/NRGRank
cd NRGRank
pip install virtualenv
virtualenv venv
source venv/bin/activate
pip install -r requirements.txt

Preparatory steps

Preprocessing the target

1. Prepare target folder:

   A folder with the target file called receptor.mol2 (must be in mol2 format; can be done in PyMol).

   A folder called get_cleft in the target folder with the desired binding site file (made with GetCleft). The name must contain "sph".

   Example:

   TARGET_FOLDER
       ├── get_cleft
       │   ├── receptor_sph_1.pdb
       ├── receptor.mol2
   

2. Preparing the target with process_target.py:

   Required Flags

   Flag	Description	Possible value(s)
   -p	Path to parent directory of the targets folder	Absolute path

   Optional Flags

   Flag	Description	Possible value(s)
   -t	Name of target folder if you want to process a specific target	Name of target
   -o	Allows overwriting existing files	N/A

   Example command to preprocess the aa2ar target in /foo/bar/targets

   python src/process_target.py -p /foo/bar/targets -t aa2ar -o
   

Preparing the ligands

All ligands for 1 screen must be in the same mol2 file. However, multiple ligand files can be prepared with one command. These can either be in one folder of multiple subdirectories of a given folder.

Required Flags

For running on a single mol2 file:

Flag	Description	Possible value(s)
-lp	Path to ligand mol2 file	Absolute path

For running on a multiple mol2 file:

Flag	Description	Possible value(s)
-fp	Path to parent directory of ligand file(s)	Absolute path

Optional Flags for running on multiple mol2 file in subdirectories

Flag	Description	Possible value(s)
-sd	For when your ligand files are in multiple subdirectories of fp	N/A

Example commands:

To preprocess the ligand file /foo/bar/ligand.mol2 file

python src/process_ligands.py -lp /foo/bar/ligand.mol2

NRGRank

You should now have a folder structure as follows and are ready to run NRGRank:

TARGET_FOLDER
    ├── get_cleft
    │   ├── receptor_sph_1.pdb
    ├── preprocessed_target
    ├── preprocessed_ligands
    ├── receptor.mol2

Example commands:

Running NRGRank on the first 1000 ligands for the target stored at /foo/bar/target:

python src/NRGRank.py -p /foo/bar/target -s 0,1001

Required Flags

Flag	Description	Possible value(s)
-p	Path to target folder	Absolute path

Optional Flags

Flag	Description	Possible value(s)
-s	Only screen subset of ligands	int,int (last is excluded)
-l	Path to preprocessed ligand folder	Absolute folder
-lt	Specify if ligand is of a special type	str
-si	Skips writing the target info file	N/A
-sr	Skips writing REMARKS before result	N/A
-o	Prevents NRGRank from creating new folders	N/A

---

Example

For this example, we will be using a target and ligands from the DUD-E database.

1. Download the target file here: https://dude.docking.org//targets/aa2ar/receptor.pdb

2. Convert the downloaded file to mol2 using a software such as PyMOL or an online converter such as the one available here: https://www.cheminfo.org/Chemistry/Cheminformatics/FormatConverter/index.html

3. Rename the target file to receptor.mol2

4. Download and extract the set of known binders here: https://dude.docking.org//targets/aa2ar/actives_final.mol2.gz

5. Make a folder (AA2AR) with the target file using the following structure:

   AA2AR
   ├── actives_final.mol2
   ├── receptor.mol2
   

6. Generate a binding site using GetCleft (1.75 minimum sphere radii / 3.75 maximum sphere radii). The easiest way is by using the NRGSuite-Qt PyMOL plugin available here: https://nrg-qt.readthedocs.io/en/latest/

7. Place the binding site file in a folder called get_cleft to obtain the following folder structure:

   AA2AR
     ├── get_cleft
     │   ├── receptor_sph_1.pdb
     ├── receptor.mol2
     ├── actives_final.mol2
   

8. Prepare the target using the following command (replace the path to the parent directory of the target):

   python src/process_target.py -p /foo/bar/targets -t AA2AR -o
   

9. Prepare the ligand file

   python src/process_ligands.py -lp /foo/bar/AA2AR/actives_final.mol2
   

10. Run NRGRank:

    python src/NRGRank.py -p /foo/bar/AA2AR -lt active