bindingsitedti-differential-scale-binding-site-modelling-for-drug-target-interaction-prediction's Introduction

BindingSiteDTI: Differential-scale binding site modelling for Drug-Target interaction prediction

Introduction

BindingSiteDTI https://doi.org/10.1093/bioinformatics/btae308 is a cutting-edge software tool developed for conducting sophisticated experiments on diverse datasets in the realm of Drug-Target Interactions (DTIs). Its primary objective is to streamline the analysis process of complex data, thereby significantly contributing to research in drug design and discovery.

Environment Setup

Prior to running BindingSiteDTI, it is essential to set up the environment by installing all necessary dependencies. This can be efficiently done through the provided requirements.txt file.

Installation Instructions

Execute the following command in your terminal to install the dependencies:

pip install -r requirements.txt

This command facilitates the automatic installation of all required packages and libraries, ensuring the software operates seamlessly.

Usage Guide

BindingSiteDTI is highly versatile, capable of supporting a variety of datasets for experimentation. The tool automatically initiates data preprocessing if processed data packages are not detected. We offer convenient one-command scripts for each experiment, as follows:

Experiment on the Human Cold Dataset

Run the following script in the terminal to conduct experiments on the human_cold dataset:

bash human.sh

Experiment on the DUDE Dataset

To perform experiments on the DUDE dataset, use this command:

bash DUDE.sh

Experiment on the BindingDB Dataset

BindingSiteDTI is equipped to handle two subsets of the BindingDB dataset. Utilize the respective scripts for these experiments:

For the BindingDB 'seen' subset:
```
bash BindingDB_seen.sh
```
For the BindingDB 'unseen' subset:
```
bash BindingDB_unseen.sh
```

We also provided a demo report for you: https://api.wandb.ai/links/panfeng-1022/7qh74d54 If you have any difficulty when running our codes, feel free to email me: [email protected]

Preprocessed Dataset (Just for saving your time)

For those who find data preprocessing to be time-consuming, we recommend downloading the preprocessed dataset from the following link:

Download Preprocessed Dataset

bindingsitedti-differential-scale-binding-site-modelling-for-drug-target-interaction-prediction's People

Contributors

Stargazers

Watchers