# Knowledge Graph Neural Network

This is our implementation for the paper

> Xuan Lin, Zhe Quan, Zhi-Jie Wang, Tengfei Ma and Xiangxiang Zeng. KGNN: Knowledge Graph Neural Network for Drug-Drug Interaction Prediction. IJCAI' 20 accepted.

<img align="center" src="Figure1.png">

Figure 1 shows the overview of KGNN. It takes the parsed DDI matrix and knowledge graph obtained from preprocessing of dataset as the input. It outputs the interaction value for the drug-drug pair. 

# Requirement

To run the code, you need the following dependencies:

* Python == 3.6.6

* Keras == 2.3.0

* Tensorflow == 1.13.1

* scikit-learn == 0.22

# Installation

You can create a virtual environment using [conda](https://github.com/conda/conda).

```bash

conda create -n kgnn python=3.6.6  

source activate kgnn  

git clone https://github.com/xzenglab/KGNN.git  

cd KGNN  

pip install -r requirement.txt  

```

# Dataset

We just provide the preprocessed KG from KEGG-drug dataset owing to the size limited. And you can directly download the original DrugBank dataset ([V5.1.4](https://www.drugbank.ca/releases/5-1-4)). Note that the construction of KG please refer to [Bio2RDF](https://github.com/bio2rdf/bio2rdf-scripts/wiki) tool in detail.

# Usage

```bash

python run.py

```

# Citation

```bash

@inproceedings{ijcai2020-380,  

  title     = {KGNN: Knowledge Graph Neural Network for Drug-Drug Interaction Prediction},  

  author    = {Lin, Xuan and Quan, Zhe and Wang, Zhi-Jie and Ma, Tengfei and Zeng, Xiangxiang},  

  booktitle = {Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence, {IJCAI-20}},  

  publisher = {International Joint Conferences on Artificial Intelligence Organization},               

  editor    = {Christian Bessiere},     

  pages     = {2739--2745},    

  year      = {2020},  

  month     = {7},  

  note      = {Main track},  

  doi       = {10.24963/ijcai.2020/380},  

  url       = {https://doi.org/10.24963/ijcai.2020/380},  

}  

```

For any clarification, comments, or suggestions please create an issue or contact [Jacklin](Jack_lin@hnu.edu.cn).