Cancer is one of the leading causes of mortality worldwide, and its inherent diversity and heterogeneity pose significant challenges in early diagnosis, drug development, and prognosis. Accurate identification of cancer subtypes has therefore become a critical aspect of personalized cancer treatment. With the advancement of omics technologies, multi-omics data offer a more comprehensive understanding of cancer's underlying mechanisms. However, effectively integrating such diverse datasets to identify cancer subtypes remains a significant challenge. In this study, we introduce a novel approach, termed HRAGNN, for cancer subtype identification through the integration of multi-omics data. HRAGNN first constructs an integrated graph and then leverages Relational Attention Mechanism (RAM) and Graph Neural Network (GNN) to capture complex features across these multi-omics layers. Subsequently, the Multi-view Fusion Network (MVFN) is employed to fuse the features derived from the different omics data. We evaluated the performance of HRAGNN on three datasets, comparing it with other existing methods. The experimental results demonstrate that HRAGNN outperforms other approaches in terms of several key evaluation metrics.

Hanwen Bai. HRAGNN: A cancer subtype identification method using multi-omics data and heterogeneous graph neural networks. Academic Journal of Computing & Information Science (2025), Vol. 8, Issue 2: 26-36. https://doi.org/10.25236/AJCIS.2025.080204.

[1] K. S. et al. Advances in cancer immunotherapy 2019 – latest trends. J. Exp. Clin. Cancer Res. 38, 268; 10.1186/s13046-019-1266-0 (2019).

[2] Alexander et al. Assessment of the molecular heterogeneity of E-Cadherin expression in invasive lobular breast cancer. Cancers 14, 295; 10.3390/cancers14020295 (2022).

[3] Ding, H. & Luo, J. MAMnet: detecting and genotyping deletions and insertions based on long reads and a deep learning approach. Brief. Bioinform. 23, bbac195; 10.1093/bib/bbac195 (2022).

[4] Hanahan, D. & Weinberg, R. A. Hallmarks of cancer: the next generation. Cell 144, 646–674 (2011).

[5] Luo, J. et al. BreakNet: detecting deletions using long reads and a deep learning approach. BMC Bioinformatics 22, 577; 10.1186/s12859-021-04499-5 (2021).

[6] S. J. P. et al. Supervised risk predictor of breast cancer based on intrinsic subtypes. J. Clin. Oncol. 41, 26 (2023).

[7] Subramanian, I., Verma, S., Kumar, S., Jere, A. & Anamika, K. Multi-omics data integration, interpretation, and its application. Bioinformatics Biol. Insights 14; 10.1177/1177932219899051 (2020).

[8] Guo, Y. et al. BCDForest: a boosting cascade deep forest model towards the classification of cancer subtypes based on gene expression data. BMC Bioinformatics 19, Suppl. 5, 118; 10.1186/s12859-018-2095-4 (2018).

[9] Zhong, L., Meng, Q., Chen, Y., Du, L. & Wu, P. A laminar augmented cascading flexible neural forest model for classification of cancer subtypes based on gene expression data. BMC Bioinformatics 22, 475; 10.1186/s12859-021-04391-2 (2021).

[10] Xu, J. et al. A novel deep flexible neural forest model for classification of cancer subtypes based on gene expression data. IEEE Access 7, 22086–22095; 10.1109/ACCESS.2019.2898723 (2019).

[11] Shen, J. et al. Deep learning approach for cancer subtype classification using high-dimensional gene expression data. BMC Bioinformatics 23, 1–17; 10.1186/s12859-022-04980-9 (2022).

[12] Cho, K., Bengio, Y. et al. Learning phrase representations using RNN encoder-decoder for statistical machine translation. In Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), 1724–1734. Association for Computational Linguistics (2014).

[13] Xu, J. et al. A hierarchical integration deep flexible neural forest framework for cancer subtype classification by integrating multi-omics data. BMC Bioinformatics 20, 527; 10.1186/s12859-019-3116-7 (2019).

[14] Wang, T. et al. MOGONET integrates multi-omics data using graph convolutional networks allowing patient classification and biomarker identification. Nature Communications 12, 3445 (2021).

[15] Wen, G. & Li, L. FGCNSurv: dually fused graph convolutional network for multi-omics survival prediction. Bioinformatics 39, 1–9; 10.1093/bioinformatics/btad472 (2023).

[16] Chen, F. et al. Supervised graph contrastive learning for cancer subtype identification through multi-omics data integration. Health Information Science and Systems 12, 12 (2024).

[17] Shen, J., Guo, X., Bai, H. & Luo, J. CAEM-GBDT: a cancer subtype identifying method using multi-omics data and convolutional autoencoder network. Frontiers in Bioinformatics 4, art. no. 1403826; 10.3389/fbinf.2024.1403826 (2024).

[18] Gao, R., Luo, J., Ding, H. & Zhai, H. INSnet: a method for detecting insertions based on deep learning network. BMC Bioinformatics 24, 80; 10.1186/s12859-023-05216-0 (2023).

[19] Fuso, A., Raia, T., Orticello, M., & Lucarelli, M. The complex interplay between DNA methylation and miRNAs in gene expression regulation. Biochimie, 173, 12-16; 10.1016/j.biochi.2020.02.006 (2020).

[20] Li, Z. Q. et al. Multi-omics analysis of five species of milk and specific composition links within each species. Food Chemistry 457, art. no. 140028; 10.1016/j.foodchem.2024.140028 (2024).

[21] Subramanian, S., Verma, S., Kumar, A., & Jere, A. K. Multi-omics data integration, interpretation, and its application. Bioinformatics and Biology Insights 14; 10.1177/1177932219899051 (2020).

[22] Su, X. et al. A comprehensive survey on community detection with deep learning. IEEE Transactions on Neural Networks and Learning Systems 35, 4, 4682–4702; 10.1109/TNNLS.2021.3137396 (2024).

[23] Hamilton, W., Ying, R., & Leskovec, J. Inductive representation learning on large graphs. In Proceedings of the Neural Information Processing Systems (2017).

[24] Mahmud, M. et al. Deep learning in mining biological data. Cognitive Computation 13, 1–33; 10.1007/s12559-020-09773-x (2021).

[25] Maghsoudi, Z., Nguyen, H., Tavakkoli, A., & Nguyen, T. A comprehensive survey of the approaches for pathway analysis using multi-omics data integration. Briefings in Bioinformatics 23, 6, art. no. bbac435; 10.1093/bib/bbac435 (2022).

[26] Ma, Y. et al. Is homophily a necessity for graph neural networks? arXiv preprint arXiv:2106.06134 (2021).