A Graph-in-Graph Learning Framework for Drug-Target Interaction Prediction

2025-07-15

AI Summary

Accurate prediction of drug–target interactions (DTIs) is critical for drug discovery, yet existing graph neural network (GNN) approaches struggle to effectively integrate cross-level features—molecular structural information and interaction topology. To address this, we propose Graph-in-Graph: a hierarchical graph framework that embeds molecular graphs of drugs and targets as meta-nodes into a global DTI interaction network, thereby unifying transductive and inductive learning within a dual-layer graph architecture. The model jointly processes SMILES strings and protein sequences to enable end-to-end, multimodal representation learning over biomedical graphs. Evaluated on a rigorously curated benchmark dataset, our method achieves state-of-the-art performance across key metrics—including AUC and AUPR—demonstrating substantial improvements in both predictive accuracy and cross-dataset generalization. Moreover, the interpretable and scalable design establishes a novel paradigm for DTI prediction, bridging structural biology with graph-based machine learning.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Reasoning under Uncertainty: Graphical Models

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 learning to rank, online learning, and counterfactual learning for ranking

Abstract

Accurately predicting drug-target interactions (DTIs) is pivotal for advancing drug discovery and target validation techniques. While machine learning approaches including those that are based on Graph Neural Networks (GNN) have achieved notable success in DTI prediction, many of them have difficulties in effectively integrating the diverse features of drugs, targets and their interactions. To address this limitation, we introduce a novel framework to take advantage of the power of both transductive learning and inductive learning so that features at molecular level and drug-target interaction network level can be exploited. Within this framework is a GNN-based model called Graph-in-Graph (GiG) that represents graphs of drug and target molecular structures as meta-nodes in a drug-target interaction graph, enabling a detailed exploration of their intricate relationships. To evaluate the proposed model, we have compiled a special benchmark comprising drug SMILES, protein sequences, and their interaction data, which is interesting in its own right. Our experimental results demonstrate that the GiG model significantly outperforms existing approaches across all evaluation metrics, highlighting the benefits of integrating different learning paradigms and interaction data.

Problem

• Predicting drug-target interactions accurately for drug discovery
• Integrating diverse drug and target features effectively
• Combining molecular and network-level interaction data

Innovation

• Graph-in-Graph model integrates molecular and network features
• Combines transductive and inductive learning for DTI prediction
• Represents molecular graphs as meta-nodes in interaction graph