Skip to content

README.md

Latest commit

 

History

History
61 lines (54 loc) · 1.74 KB

README.md

File metadata and controls

61 lines (54 loc) · 1.74 KB
Raw

Docked Compound Cluster Visualiser

This application provides a GUI for clustering and visualising docked compounds based on molecular descriptors and fingerprints using DBSCAN. It includes a 3D viewer for selected molecules.

Executable

Requirement

A valid OpenEye license.

Steps to run the executable

  1. Set the environment variable:
%SystemRoot%\System32\setx OE_LICENSE "C:\path\to\oe_license.txt"
  1. Double-click the executable.

Script

Python version

Python 3.9

Dependencies

OpenEye Python Toolkit is not included here because it must be downloaded manually from https://openeye.app.box.com/s/ebywkngy0p45enmn3za21kb7dzf1mzbl/folder/306164025307, and requires a valid license.

A few Python packages need to be installed.

Steps to run the script

  1. Create a new virtual environment:
python -m venv env
env\Scripts\activate
  1. Install the wheel:
pip install "C:\path\to\OpenEye_toolkits-<version>-py39-none-<platform>.whl"
  1. Set the environment variable:
%SystemRoot%\System32\setx OE_LICENSE "C:\path\to\oe_license.txt"
  1. Install the required packages:
pip install -r "C:\path\to\requirements.txt"
  1. If requirements.txt is not working, install the packages manually:
pip install kneed matplotlib "numpy<2" pandas "pillow>=10,<11" py3Dmol rdkit-pypi scikit-learn ttkbootstrap
  1. Run the script:
python "C:\path\to\ClusteringApp.py"

Note

This application requires an input file in .oeb or .oeb.gz format that contains the following keywords so it can proceed with clustering:

  • title
  • donor
  • acceptor
  • rotatable
  • logp
  • tpsa
  • mw
  • smiles
  • fred chemgauss4 score