BOMIFA: biologically informed multi-omics integration with graph contrastive learning for cancer prognosis in women
Briefings in Bioinformatics · 2026
DOI: 10.1093/bib/bbag500 ↗Kurz gefasst
Fachartikel in der Zeitschrift Briefings in Bioinformatics, 2026. Verfasst von Zixiao Lu und 8 weiteren Autor:innen. Beteiligte Einrichtung: School of Medical Information Engineering, Guangzhou University of Chinese Medicine, No. 232 Waihuan East Road, Panyu District, Guangzhou 510006, Chi…. Hauptthemen laut MEDLINE: Biomarkers, Tumor, Neoplasms, Computational Biology. Die Arbeit nennt 49 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: National Natural Science Foundation of China, Guangdong Provincial Medical Research Foundation, Guangdong Provincial Administration of Traditional Chinese Medicine TCM Research Projects.
Abstract (Originalsprache)
Accurate survival prediction remains a central challenge in precision oncology, particularly for female patients whose sex-specific molecular characteristics are often under-modeled in prior studies. Although multi-omics integration enables a deeper exploration of prognostic biomarkers, existing methods rely on mathematically driven fusion strategies, which tend to dilute omics-specific signals and fail to capture biological regulatory hierarchies across omics layers. To address these limitations, we propose BOMIFA (Biologically informed Omics representation and Multi-omics Integration Framework), a deep graph-based framework for survival prediction and biomarker discovery in female patients using DNA methylation, mRNA, and miRNA expression data. BOMIFA incorporates two key innovations. First, graph contrastive learning is leveraged within each omics encoder to enhance intra-omics representation learning and amplify prognostically relevant signals. Then, a biologically informed cross-omics attention mechanism is deployed to explicitly model directional regulatory dependencies, enabling inter-omics information exchange aligned with known molecular hierarchies. Extensive benchmarking on eight cancer cohorts demonstrates that BOMIFA consistently outperforms existing prognostic methods in female patients. Moreover, saliency map-based gradient attribution enables the identification of female-associated prognostic biomarkers that were overlooked in prior mixed-sex analyses.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Substanzen
Autor:innen und Institutionen
- School of Medical Information Engineering, Guangzhou University of Chinese Medicine, No. 232 Waihuan East Road, Panyu District, Guangzhou 510006, China.
- School of Medical Information and Engineering, Guangdong Pharmaceutical University, No. 280 Waihuan East Road, Panyu District, Guangzhou 510006, China.
- School of Life Sciences and Biopharmaceutics, Guangdong Pharmaceutical University, No. 280 Waihuan East Road, Panyu District, Guangzhou 510006, China.
- Research Center for Ubiquitous Computing Systems, Institute of Computing Technology, Chinese Academy of Sciences, No. 6 Kexueyuan South Road, Haidian District, Beijing 100190, China.
- Guangdong Province Precise Medicine Big Data of Traditional Chinese Medicine Engineering Technology Research Center, Guangdong Pharmaceutical University, No. 280 Waihuan East Road, Panyu District, Guangzhou 510006, Chin…
Förderung
- National Natural Science Foundation of China
- Guangdong Provincial Medical Research Foundation
- Guangdong Provincial Administration of Traditional Chinese Medicine TCM Research Projects
Zitieren
APA
Lu, Z., Wang, J., Liang, Y., Lin, Z., Tan, Y., Ma, Q., Zhou, W., Zhao, Y., & Xu, S. (2026). BOMIFA: biologically informed multi-omics integration with graph contrastive learning for cancer prognosis in women. Briefings in Bioinformatics. https://doi.org/10.1093/bib/bbag500
Vancouver
Lu Z, Wang J, Liang Y, Lin Z, Tan Y, Ma Q, et al. BOMIFA: biologically informed multi-omics integration with graph contrastive learning for cancer prognosis in women. Briefings in Bioinformatics. 2026. doi:10.1093/bib/bbag500
BibTeX
@article{Lu2026,
author = {Lu, Zixiao and Wang, Jiajun and Liang, Yuping and Lin, Zhenghao and Tan, Yingyin and Ma, Qian and Zhou, Wu and Zhao, Yi and Xu, Siwen},
title = {BOMIFA: biologically informed multi-omics integration with graph contrastive learning for cancer prognosis in women},
journal = {Briefings in Bioinformatics},
year = {2026},
doi = {10.1093/bib/bbag500}
}
EUVIMED