Hierarchical graph neural network with subgraph perturbations for key gene cluster discovery in cancer staging
Open Access
- 17 July 2023
- journal article
- research article
- Published by Springer Nature in Complex & Intelligent Systems
- Vol. 10 (1) , 111-128
Abstract
Analyzing highly individual-specific genomic data to understand genetic interactions in cancer development is still challenging, with significant implications for the discovery of individual biomarkers as well as personalized medicine. With the rapid development of deep learning, graph neural networks (GNNs) have been employed to analyze a wide range of biomolecular networks. However, many neural networks are limited to black box models, which are only capable of making predictions, and they are often challenged to provide reliable biological and clinical insights. In this research, for sample-specific networks, a novel end-to-end hierarchical graph neural network with interpretable modules is proposed, which learns structural features at multiple scales and incorporates a soft mask layer in extracting subgraphs that contribute to classification. The perturbations caused by the input graphs' deductions are used to evaluate key gene clusters, and the samples are then grouped into classes to produce both sample- and stage-level explanations. Experiments on four gene expression datasets from The Cancer Genome Atlas (TCGA) show that the proposed model not only rivals the advanced GNN methods in cancer staging but also identifies key gene clusters that have a great impact on classification confidence, providing potential targets for personalized medicine.Keywords
Funding Information
- National Natural Science Foundation of China (62072212)
- Jilin Provincial Scientific and Technological Development Program (20220508125RC, 20210508060RQ)
- National Key R&D Program (2018YFC2001302)
- Jilin Provincial Key Laboratory of Big Data Intelligent Cognition (20210504003GH)
This publication has 73 references indexed in Scilit:
- Cancer driver gene discovery through an integrative genomics approach in a non-parametric Bayesian frameworkBioinformatics, 2016
- The expression profile and clinic significance of the SIX family in non-small cell lung cancerJournal of Hematology & Oncology, 2016
- Nerve growth factor (NGF)-mediated regulation of p75NTR expression contributes to chemotherapeutic resistance in triple negative breast cancer cellsBiochemical and Biophysical Research Communications, 2016
- Primary breast cancer induces pulmonary vascular hyperpermeability and promotes metastasis via the VEGF–PKC pathwayMolecular Carcinogenesis, 2015
- Clonal status of actionable driver events and the timing of mutational processes in cancer evolutionScience Translational Medicine, 2015
- Cell–cell adhesion genes CTNNA2 and CTNNA3 are tumour suppressors frequently mutated in laryngeal carcinomasNature Communications, 2013
- The Association of Intratumoral Germinal Centers with Early-Stage Non-small Cell Lung CancerJournal of Thoracic Oncology, 2011
- The CCL21/CCR7 pathway plays a key role in human colon cancer metastasis through regulation of matrix metalloproteinase-9Digestive and Liver Disease, 2011
- Characterization of the roles of RHOC and RHOA GTPases in invasion, motility, and matrix adhesion in inflammatory and aggressive breast cancersCancer, 2010
- Systematic and integrative analysis of large gene lists using DAVID bioinformatics resourcesNature Protocols, 2008