What question did this study set out to answer?

The aim is to enhance the classification of cancer subtypes using multi-omics data through a novel integrated graph convolutional network model.

February 28, 2026Open Access

Robust graph structure learning to improve multi-omics cancer subtype classification

Key Points

The aim is to enhance the classification of cancer subtypes using multi-omics data through a novel integrated graph convolutional network model.
Developed the FaGGCN model combining feature and graph structure learning.
Utilized convolutional autoencoders for latent feature extraction.
Leverage patient survival information to identify significant features.
Fuse key features with inter-omics similarity matrices for comprehensive network learning.
Conducted survival, sensitivity, and differential gene expression analyses to interpret model outcomes.
Achieved competitive classification performance across eight cancer datasets.
Demonstrated improved classification accuracy and exploratory survival predictions.
Highlighted interpretable biomarkers important for clinical applications.

Abstract

Classifying cancer patients into consistent subtypes at the multi-omics level remains a significant challenge in advancing precision medicine. Nevertheless, a key problem in integrating multi-omics data lies in concurrently addressing intra-omics and inter-omics information, along with sample networks. In this study, we introduce the Feature and Graph Structure-Learning Integrated Graph Convolutional Network (FaGGCN), which combines feature learning and graph structure learning for multi-omics cancer subtyping. The model employs convolutional autoencoders to learn information-rich latent features, and patient survival information is further leveraged to select key features that are significantly associated with survival outcomes. The graph autoencoder fuses the key features with inter-omics similarity fusion matrices, enabling the model to learn a comprehensive sample network. Finally, the graph convolutional network integrates the key features while incorporating the sample network to precisely classify patients. Additionally, survival analysis, sensitivity analysis, and differential gene expression analysis highlight the interpretability of the FaGGCN model, as well as its ability to identify biomarkers suitable for clinical research. Experimental results show that our model achieves competitive performance across eight cancer datasets spanning four omics modalities, with generally improved classification performance and exploratory survival prediction results.

Mark Helpful

Bookmark

Relay

View Full Paper