EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

M-GNN: A Topology-Enhanced Multi-Modal Graph Neural Network for Cancer Driver Gene Prediction

Lu Qin iD, Wen Zhu, Xinyi Liao, Yujing Zhang

Metabolites · 16.04.2026 · 16 (4) · 268

DOI: 10.3390/metabo16040268 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Metabolites, online erschienen am 16.04.2026. Verfasst von Lu Qin und 3 weiteren Autor:innen. Beteiligte Einrichtung: School of Mathematics and Statistics, Hainan Normal University, Haikou 571158, China. Die Arbeit nennt 56 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, National Key R and D Program of China, Natural Science Foundation of Hainan, China.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Accurate identification of cancer driver genes is essential for understanding tumorigenesis and developing targeted therapies. Although graph neural networks (GNNs) have advanced multi-omics integration, existing methods often simply concatenate omics features and underutilize the topological information of biological networks.

Methods

We propose M-GNN, a multi-modal GNN framework for cancer driver gene prediction. It employs separate Graph Convolutional Network (GCN) encoders to process four types of omics data (mutation, expression, methylation, copy number variation (CNV)), each represented as a 16-dimensional vector. We incorporate knowledge distillation by using soft labels from a pre-trained teacher model to enhance feature representation. An attention mechanism adaptively fuses the encoded omics features, and a dual-path classifier combining a GCN and a Multilayer Perceptron (MLP) preserves both intrinsic gene properties and network topology.

Results

Experiments on three public protein–protein interaction (PPI) networks show that M-GNN consistently achieves the highest or second-highest AUPRC compared to five state-of-the-art methods. Ablation studies confirm the contribution of each module, and biological interpretability analysis—including analysis of GO enrichment and drug sensitivity—validates the reliability of the predicted genes.

Conclusions

M-GNN provides a robust and interpretable computational tool for systematic cancer driver gene identification, effectively integrating multi-omics and network data.

Schlagwörter

Stichwörter der Autor:innen

cancer driver genesgraph neural networksknowledge distillationmulti-omics integrationtopological feature enhancement

Autor:innen und Institutionen

  • School of Mathematics and Statistics, Hainan Normal University, Haikou 571158, China.
  • Key Laboratory of Data Science and Intelligence Education, Hainan Normal University, Ministry of Education, Haikou 571158, China.
  • School of Mathematics and Systems Science, Guangdong Polytechnic Normal University, Guangzhou 510665, China.
  • China Unicom (Hainan) Industrial Internet Co., Ltd., Haikou 571924, China.

ORCID-Profile: Lu Qin

Förderung

  • National Natural Science Foundation of China
  • National Key R and D Program of China
  • Natural Science Foundation of Hainan, China
  • Program of Graduate 497 Education and Teaching Reform in Hainan, China
  • Program of Graduate Education and Teaching Reform in Hainan, China

Zitieren

APA

Qin, L., Zhu, W., Liao, X., & Zhang, Y. (2026). M-GNN: A Topology-Enhanced Multi-Modal Graph Neural Network for Cancer Driver Gene Prediction. Metabolites, 16(4), 268. https://doi.org/10.3390/metabo16040268

Vancouver

Qin L, Zhu W, Liao X, Zhang Y. M-GNN: A Topology-Enhanced Multi-Modal Graph Neural Network for Cancer Driver Gene Prediction. Metabolites. 2026;16(4):268. doi:10.3390/metabo16040268

BibTeX
@article{Qin2026,
  author = {Qin, Lu and Zhu, Wen and Liao, Xinyi and Zhang, Yujing},
  title = {M-GNN: A Topology-Enhanced Multi-Modal Graph Neural Network for Cancer Driver Gene Prediction},
  journal = {Metabolites},
  year = {2026},
  volume = {16},
  number = {4},
  pages = {268},
  doi = {10.3390/metabo16040268}
}
Kostenlos starten →