A Novel Method for Prediction of Druggable RNA-Small Molecule Binding Pockets Using Machine Learning

Sowmya Ramaswamy Krishnan, Arijit Roy^#, Limsoon Wong, M. Michael Gromiha^#

NOTE: This code should be used for academic purposes only. This code should not be shared or used for commercial purposes without the consent of all the authors involved.

Description: Python implementation of the programs used for development of the models hosted in Druggable RNA-Ligand binding Pocket Selector (DRLiPS) web server

Usage restrictions

The code has been provided for academic purposes only.

Data

The sample datasets used for training and evaluating the models are provided in the data folder. In order to use your own data or the complete dataset, you have to prepare the datasets in the format provided for each case to train/test the models. To get the complete datasets used in the study, please visit the PDB database.

Dependencies not covered

The feature calculations for binding sites require the X3DNA-DSSR binary executable. Since it is a third-party licensed tool, it will not be included here. Kindly ensure that the executable is available in your system before running the code.

Code usage

Instructions on code usage are available in a README.md file within each folder (links also provided below).

• Dataset pre-processing
• Feature selection
• Model evaluation

For further queries related to code usage, please write to us ([email protected]; [email protected]; [email protected]).

Citation

Please cite this article if you use the codes in this repository for your research: