Transferable Binding Principles Meta-Learnig for Cross-Domain Drug-Target Interaction Prediction

This repository contains the implementation of the meta-learning framework described in the paper "Transferable Binding Principles Meta-Learnig for Cross-Domain Drug-Target Interaction Prediction".

Table of Contents

• Requirements
• Installation
• Running the Code
  • Regular Learning Mode
  • Running with CADA Module
• Citation
• License

Requirements

• Python 3.8 or higher
• PyTorch (compatible with your CUDA version, if using GPU)
• Learn2Learn
• PyTorch Lightning
• And other dependencies listed in requirements.txt

Installation

To set up the environment, follow these steps:

1. Clone the repository:

   git clone https://github.com/lian-xiao/BioBridge.git
   cd BioBridge

2. Create a virtual environment (optional but recommended):

   python -m venv venv
   source venv/bin/activate  # On Windows use `venv\Scripts\activate`

3. Install the dependencies:

   pip install -r requirement.txt

Running the Code

Regular Learning Mode

To run the regular learning mode, execute the following command:

python Dti_cnn_main.py

This script will train the model using the regular learning approach as described in the paper.

Running with CADA Module

To run the script using the CADA module for cross-domain drug-target interaction prediction, execute the following command:

python CADA/main.py

This script will utilize the CADA module to assist in predicting drug-target interactions across different domains, as detailed in the paper.