nrgrank 1.0.14

A Python based virtual screening software!

Installation (Requires Python 3.9 and above)

Use the following command:

pip install nrgrank

To install nrgrank with rdkit for generating conformers:

pip install "nrgrank[rdkit]"

Using NRGRank in a python script:

from nrgrank import process_target, process_ligands, nrgrank_main

Preparatory steps

1. Preprocessing the target

Required Arguments

Argument	Description	Possible value(s)
target_mol2_path	Path to target mol2 file	Absolute path
binding_site_file_path	Path to binding site file generated by GetCleft	Absolute path

Optional Arguments

Flag	Description	Possible value(s)
create_new_dir	A new folder will be created to store generated files	True,False
overwrite	Allows overwriting existing files	True,False

Example command

target_save_dir = process_target(target_mol2_path='foo/bar/target.mol2', binding_site_file_path='foo/bar/bd_site_sph_1.pdb')

2. Preparing the ligands

All ligands for one screen must be in the same mol2 file

Required Arguments

For running on a single mol2 file:

Flag	Description	Possible value(s)
ligand_path	Path to ligand mol2 file	Absolute path
conformers_per_molecule	Number of conformers per molecule	Int

Optional Arguments

Flag	Description	Possible value(s)
overwrite	Allows overwriting existing files	True,False
ligand_type	Assign a specific ligand type recognised by NRGRank. Can be useful for benchmarking.	(str)
output_dir	Folder where generated files will be stored	Absolute path

Example commands:

ligand_save_dir = process_ligands(ligand_path='foo/bar/ligands.mol2', conformers_per_molecule=1)

---

NRGRank

Required Flags

Flag	Description	Possible value(s)
target_name	The name of the target. Used for naming the output CSV file when csv_and_pose_name is not specified	(str)
preprocessed_target_path	Path to the preprocessed target folder (e.g., output from process_target())	Absolute path
preprocessed_ligand_path	Path to the preprocessed ligand folder (e.g., output from process_ligand())	Absolute path
result_folder_path	Directory path where the docking results will be stored	Absolute path

---

Optional Flags

Flag	Description	Possible value(s)
result_csv_and_pose_name	Custom name for CSV and pose files. If not given, defaults to target_name	(str)
ligand_type	Type of ligand being processed. Useful when benchmarking with DUD-E	"ligand", "active", "decoy"
ligand_slice	Slice range of ligands to process	list[int] or None
write_info	Write informational remarks to output (e.g., .txt log)	True, False
write_file	Output result file(s)	True, False
output_dictionary	Output results as a Python dictionary	True, False
file_separator	Separator character for the result file (CSV/TSV)	",", "\t"
output_header	Write a header row in the CSV file	True, False
save_time	Save time taken per molecule for docking	True, False
normalise_score	Normalise docking score by number of atoms in molecule	True, False
unique_run_id	Unique identifier to avoid file name conflicts and used to name ligand poses	(str)
user_config	Arbitrary keyword arguments for overriding default docking parameters	key=value pairs

Example commands:

result_file_path, result_csv_lines = nrgrank_main(target_name='target', 
                   preprocessed_target_path='foo/bar/preprocessed_target', 
                   preprocessed_ligand_path='foo/bar/preprocessed_ligands_1_conf',
                   result_folder_path='foo/bar/results')

Returns the path to the result file and the lines contained in the result csv