Heterogeneous networks in drug-target interaction prediction

2025-04-22

AI Summary

Drug–target interaction (DTI) prediction remains a critical bottleneck in drug discovery, hindered by the high cost and long turnaround time of wet-lab experiments. This work presents a systematic review of 32 state-of-the-art DTI prediction studies published between 2020 and 2024 that leverage heterogeneous graph neural networks (HGNNs). We establish a structured analytical framework covering architecture design, multimodal graph representations (e.g., molecular graphs for drugs; sequence- or structure-based graphs for targets), cross-domain node alignment strategies, benchmark datasets (BindingDB, DrugBank, Davis), and evaluation metrics. Furthermore, we propose reproducibility guidelines and standardized evaluation protocols tailored to biomedical graph learning, alongside an open-source meta-database and code index. Under unified experimental settings, we benchmark over 15 HGNN-based models. Our contributions provide a methodological foundation and technical infrastructure for interpretable DTI modeling and multi-center validation.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

Drug discovery requires a tremendous amount of time and cost. Computational drug-target interaction prediction, a significant part of this process, can reduce these requirements by narrowing the search space for wet lab experiments. In this survey, we provide comprehensive details of graph machine learning-based methods in predicting drug-target interaction, as they have shown promising results in this field. These details include the overall framework, main contribution, datasets, and their source codes. The selected papers were mainly published from 2020 to 2024. Prior to discussing papers, we briefly introduce the datasets commonly used with these methods and measurements to assess their performance. Finally, future challenges and some crucial areas that need to be explored are discussed.

Problem

• Predicting drug-target interactions using heterogeneous networks
• Reducing time and cost in drug discovery via computational methods
• Surveying graph machine learning approaches for interaction prediction

Innovation

• Graph machine learning for drug-target prediction
• Comprehensive review of 2020-2024 methods
• Includes datasets and performance metrics