EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

TMO-Net+: An Enhanced Tumor Multi-Omics Pre-Trained Network for Multi-Task Learning in Oncology

Wei Liu iD, Xuan Liu iD, Shuyu Zhou, Kaiyang Li, Xiangzhi Wang, Ke Chen, Lilu Guo iD, Rui Zhang, Qingzhi Su

Genes · 09.09.2026 · 17 (9) · 1085

DOI: 10.3390/genes17091085 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Genes, online erschienen am 09.09.2026. Verfasst von Wei Liu und 8 weiteren Autor:innen. Beteiligte Einrichtung: School of Computer and Software, Nanyang Institute of Technology, Changjiang Road 80, Nanyang 473000, China. Hauptthemen laut MEDLINE: Neoplasms. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 49 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Henan Provincial Science and Technology Research Project, Nanyang Basic and Frontier Research Project, Nanyang Science and Technology Research Project.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Tumor heterogeneity arises from complex interactions among diverse biological factors, posing a major challenge for the development of robust multi-omics data integration methods. While the existing Tumor Multi-Omics pre-trained Network (TMO-Net) enables the fusion of multi-omics features into unified representations, its practical utility is constrained by issues such as missing modalities, incomplete within-omics data, and high-dimensional noise. To overcome these limitations, we propose TMO-Net+, an enhanced architecture specifically designed to improve the robustness and reliability of multi-omics modeling.

Methods

TMO-Net+ introduces several coordinated architectural enhancements. First, a feature attention encoder is applied to each omics data type to reduce the influence of modality-dependent input variation. Second, we combine a gated Mixture-of-Experts (MoE) module with a Product-of Experts (PoE) mechanism to capture sample-specific contributions and enable robust inference even when partial omics data are available. Additionally, a supervised deep classification head with a tailored loss function is incorporated to enhance the separability of learned embeddings in the latent space.

Results

Extensive experiments on pan-cancer datasets demonstrate that TMO-Net+ consistently outperforms the original TMO-Net, as measured by LogME scores. Furthermore, in various downstream tasks (e.g., pan-cancer classification, primary/metastatic site prediction, and prognostic modeling), TMO-Net+ achieves superior performance under partial-omics settings, which proves that it enhances the robustness and cross-cancer transferability of the multi-omics representations.

Conclusions

The proposed TMO-Net+ improves the robustness and cross-cancer transferability of multi-omics representations within the evaluated TCGA cohorts. Biological interpretability analyses further show that TMO-Net+ prioritizes established cancer-driver genes, preserves cancer-dependent molecular-state information, and adaptively redistributes relative modality contributions across molecular states. By addressing modality-level missingness and modality-dependent input variation, it offers a reliable framework for integrative tumor analysis within the evaluated TCGA cohorts.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

NeoplasmsHumansMultiomicsDeep Learning

Stichwörter der Autor:innen

TMO-Net+classifier lossfeature attention encodergated Mixture-of-Expertstumor prediction

Autor:innen und Institutionen

  • School of Computer and Software, Nanyang Institute of Technology, Changjiang Road 80, Nanyang 473000, China.
  • Academy for Electronic Information Discipline Studies, Nanyang Institute of Technology, Changjiang Road 80, Nanyang 473000, China.
  • Nanyang Central Hospital, Gongnong Road 312, Nanyang 473000, China.
  • Xinye County Xinruo Electronic Technology Co., Ltd., No. 009 Xingye Road, Xinye County, Nanyang 473000, China.

ORCID-Profile: Wei Liu, Xuan Liu, Lilu Guo

Förderung

  • Henan Provincial Science and Technology Research Project
  • Nanyang Basic and Frontier Research Project
  • Nanyang Science and Technology Research Project
  • Interdisciplinary Sciences Project, Nanyang Institute of Technology
  • Interdisciplinary Sciences Project
  • Nanyang Institute of Technology

Zitieren

APA

Liu, W., Liu, X., Zhou, S., Li, K., Wang, X., Chen, K., Guo, L., Zhang, R., & Su, Q. (2026). TMO-Net+: An Enhanced Tumor Multi-Omics Pre-Trained Network for Multi-Task Learning in Oncology. Genes, 17(9), 1085. https://doi.org/10.3390/genes17091085

Vancouver

Liu W, Liu X, Zhou S, Li K, Wang X, Chen K, et al. TMO-Net+: An Enhanced Tumor Multi-Omics Pre-Trained Network for Multi-Task Learning in Oncology. Genes. 2026;17(9):1085. doi:10.3390/genes17091085

BibTeX
@article{Liu2026,
  author = {Liu, Wei and Liu, Xuan and Zhou, Shuyu and Li, Kaiyang and Wang, Xiangzhi and Chen, Ke and Guo, Lilu and Zhang, Rui and Su, Qingzhi},
  title = {TMO-Net+: An Enhanced Tumor Multi-Omics Pre-Trained Network for Multi-Task Learning in Oncology},
  journal = {Genes},
  year = {2026},
  volume = {17},
  number = {9},
  pages = {1085},
  doi = {10.3390/genes17091085}
}
Kostenlos starten →