TF-DWGNet: A Directed Weighted Graph Neural Network with Tensor Fusion for Multi-Omics Cancer Subtype Classification

2025-09-19

AI Summary

Existing methods for multi-omics cancer subtyping struggle with high data heterogeneity, high dimensionality, and complex inter-modal dependencies, failing to effectively model the directionality and strength of biological interactions while lacking interpretability. To address these challenges, we propose TF-DWGNet: (1) a supervised tree-based approach constructs modality-specific directed weighted graphs to explicitly encode intra- and inter-omic dependency directions and strengths; (2) a low-rank decomposition-based tensor fusion mechanism jointly models uni-, bi-, and tri-modal interactions, enabling end-to-end joint embedding and feature-level interpretability. Evaluated on multiple TCGA cancer cohorts, TF-DWGNet achieves statistically significant performance improvements over state-of-the-art methods (p < 0.01). Biological validation confirms its ability to accurately identify driver genes, dysregulated pathways, and dominant omics modalities, demonstrating both predictive accuracy and mechanistic interpretability.

Technology Category

• Machine Learning: Multimodal Learning
• Computer Vision: Multi-modal Vision
• Knowledge Representation and Reasoning: Ontologies

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in 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: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

Integration and analysis of multi-omics data provide valuable insights for cancer subtype classification. However, such data are inherently heterogeneous, high-dimensional, and exhibit complex intra- and inter-modality dependencies. Recent advances in graph neural networks (GNNs) offer powerful tools for modeling such structure. Yet, most existing methods rely on prior knowledge or predefined similarity networks to construct graphs, which are often undirected or unweighted, failing to capture the directionality and strength of biological interactions. Interpretability at both the modality and feature levels also remains limited. To address these challenges, we propose TF-DWGNet, a novel Graph Neural Network framework that combines tree-based Directed Weighted graph construction with Tensor Fusion for multiclass cancer subtype classification. TF-DWGNet introduces two key innovations: a supervised tree-based approach for constructing directed, weighted graphs tailored to each omics modality, and a tensor fusion mechanism that captures unimodal, bimodal, and trimodal interactions using low-rank decomposition for efficiency. TF-DWGNet enables modality-specific representation learning, joint embedding fusion, and interpretable subtype prediction. Experiments on real-world cancer datasets show that TF-DWGNet consistently outperforms state-of-the-art baselines across multiple metrics and statistical tests. Moreover, it provides biologically meaningful insights by ranking influential features and modalities. These results highlight TF-DWGNet's potential for effective and interpretable multi-omics integration in cancer research.

Problem

• Classifying cancer subtypes using heterogeneous multi-omics data
• Capturing directional and weighted biological interactions in data
• Improving interpretability for modality and feature-level insights

Innovation

• Tree-based directed weighted graph construction for each omics modality
• Tensor fusion with low-rank decomposition for multimodal interactions
• Interpretable feature and modality ranking for biological insights