Heterogeneous Graph based Deep Learning for Biomedical Network Link Prediction

TL;DR

This paper introduces GPLP, a graph neural network model that predicts biomedical network links using topological data, demonstrating superior performance on multiple heterogeneous biomedical networks and robustness to incomplete data.

Contribution

The paper presents a novel GNN-based link prediction model, GPLP, specifically designed for heterogeneous biomedical networks, with improved accuracy and robustness over existing methods.

Findings

GPLP outperforms state-of-the-art baselines in biomedical link prediction.

The method is effective across different types of biomedical networks.

GPLP maintains robustness even with incomplete network data.

Abstract

Multi-scale biomedical knowledge networks are expanding with emerging experimental technologies that generates multi-scale biomedical big data. Link prediction is increasingly used especially in bipartite biomedical networks to identify hidden biological interactions and relationshipts between key entities such as compounds, targets, gene and diseases. We propose a Graph Neural Networks (GNN) method, namely Graph Pair based Link Prediction model (GPLP), for predicting biomedical network links simply based on their topological interaction information. In GPLP, 1-hop subgraphs extracted from known network interaction matrix is learnt to predict missing links. To evaluate our method, three heterogeneous biomedical networks were used, i.e. Drug-Target Interaction network (DTI), Compound-Protein Interaction network (CPI) from NIH Tox21, and Compound-Virus Inhibition network (CVI). Our proposed GPLP method significantly outperforms over the state-of-the-art baselines. In addition, different network incompleteness is analysed with our devised protocol, and we also design an effective approach to improve the model robustness towards incomplete networks. Our method demonstrates the potential applications in other biomedical networks.