Hierarchical Pooling and Explainability in Graph Neural Networks for Tumor and Tissue-of-Origin Classification Using RNA-seq Data

2026-01-10

arXiv.org

AI Summary

This study addresses the challenge of accurately distinguishing tumor from normal samples and identifying their tissue of origin using RNA-seq data, while enhancing model interpretability. We propose a novel interpretable hierarchical graph neural network that integrates TCGA gene expression profiles with the STRING protein–protein interaction network. By employing Chebyshev graph convolution (K=2) and weighted pooling, genes are clustered into supernodes that preserve critical biological interaction information during dimensionality reduction. Gradient-based saliency analysis is leveraged to identify driver genes and enriched pathways. Experimental results demonstrate that a single-layer pooling architecture achieves optimal performance (macro-F1 of 0.978), effectively mitigating over-smoothing and successfully recovering known cancer-associated genes and pathways, thereby offering a new paradigm for biomarker discovery.

Technology Category

Machine Learning: Graph-based Machine LearningNatural Language Processing: Interpretability, Analysis, and Evaluation of NLP ModelsData Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community

Application Category

Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphsSearch and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for rankingSemantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics

Abstract

This study explores the use of graph neural networks (GNNs) with hierarchical pooling and multiple convolution layers for cancer classification based on RNA-seq data. We combine gene expression data from The Cancer Genome Atlas (TCGA) with a precomputed STRING protein-protein interaction network to classify tissue origin and distinguish between normal and tumor samples. The model employs Chebyshev graph convolutions (K=2) and weighted pooling layers, aggregating gene clusters into'supernodes'across multiple coarsening levels. This approach enables dimensionality reduction while preserving meaningful interactions. Saliency methods were applied to interpret the model by identifying key genes and biological processes relevant to cancer. Our findings reveal that increasing the number of convolution and pooling layers did not enhance classification performance. The highest F1-macro score (0.978) was achieved with a single pooling layer. However, adding more layers resulted in over-smoothing and performance degradation. However, the model proved highly interpretable through gradient methods, identifying known cancer-related genes and highlighting enriched biological processes, and its hierarchical structure can be used to develop new explainable architectures. Overall, while deeper GNN architectures did not improve performance, the hierarchical pooling structure provided valuable insights into tumor biology, making GNNs a promising tool for cancer biomarker discovery and interpretation

Problem

• Graph Neural Networks
• RNA-seq
• Tumor Classification
• Tissue-of-Origin
• Explainability

Innovation

• Hierarchical Pooling
• Graph Neural Networks
• Explainability
• Chebyshev Graph Convolutions
• RNA-seq