Multi-omics survival prediction under modality incompleteness via dynamic modality graphs and product-of-experts fusion
Briefings in Bioinformatics · 02.11.2026 · 27 (6)
DOI: 10.1093/bib/bbag596 ↗Kurz gefasst
Fachartikel in der Zeitschrift Briefings in Bioinformatics, online erschienen am 02.11.2026. Verfasst von Weidong Xie und 4 weiteren Autor:innen. Beteiligte Einrichtung: School of Computer Science and Engineering, Northeastern University, 195 Chuangxin Road, Hunnan District, Shenyang, Liaoning 110169, China. Hauptthemen laut MEDLINE: Neoplasms, Computational Biology. Die Arbeit nennt 43 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: National Natural Science Foundation of China, Natural Science Foundation of Liaoning Province, National Frontiers Science Center for Industrial Intelligence and Systems Optimization under the 111 Project.
Abstract (Originalsprache)
Multi-omics data integration provides rich complementary information for cancer survival prediction, yet modality incompleteness is pervasive in real-world medical data and undermines the stability and applicability of existing models. To address this challenge, we propose DMG-PoE, a robust learning framework for multi-omics survival prediction. The framework introduces a dynamic modality graph (DMG) that propagates complementary information across modality nodes to compensate for missing inputs, combined with an uncertainty-aware product-of-experts (PoE) fusion mechanism that adaptively weights each modality by its estimated precision, and a consistency-constrained training strategy that stabilizes predictions across varying modality subsets. Systematic evaluation on the SurvBoard benchmark across 10 TCGA cancer types and seven omics modalities demonstrates that DMG-PoE achieves the highest average C-index under both the standard setting (complete-modality training/testing) and the missing-augmented setting (training additionally includes real incomplete samples and nonclinical modality dropout, with complete-only testing); controlled per-modality masking experiments validate effective test-time recovery in 22/24 cancer-modality combinations, and cross-cohort transfer with only two of seven modalities available further confirms generalization under severe input restriction.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- School of Computer Science and Engineering, Northeastern University, 195 Chuangxin Road, Hunnan District, Shenyang, Liaoning 110169, China.
- Key Laboratory of Intelligent Computing in Medical Image (MIIC), Northeastern University, No. 11, Lane 3, Wenhua Road, Heping District, Shenyang, Liaoning 110819, China.
- National Frontiers Science Center for Industrial Intelligence and Systems Optimization, Northeastern University, No. 11, Lane 3, Wenhua Road, Heping District, Shenyang, Liaoning 110819, China.
- College of Medicine and Bioinformation Engineering, Northeastern University, 195 Chuangxin Road, Hunnan District, Shenyang, Liaoning 110169, China.
ORCID-Profile: Weidong Xie, Jinzhu Yang, Kun Yu, Wei Li
Förderung
- National Natural Science Foundation of China
- Natural Science Foundation of Liaoning Province
- National Frontiers Science Center for Industrial Intelligence and Systems Optimization under the 111 Project
Zitieren
APA
Xie, W., Zhang, H., Yang, J., Yu, K., & Li, W. (2026). Multi-omics survival prediction under modality incompleteness via dynamic modality graphs and product-of-experts fusion. Briefings in Bioinformatics, 27(6). https://doi.org/10.1093/bib/bbag596
Vancouver
Xie W, Zhang H, Yang J, Yu K, Li W. Multi-omics survival prediction under modality incompleteness via dynamic modality graphs and product-of-experts fusion. Briefings in Bioinformatics. 2026;27(6). doi:10.1093/bib/bbag596
BibTeX
@article{Xie2026,
author = {Xie, Weidong and Zhang, Huixia and Yang, Jinzhu and Yu, Kun and Li, Wei},
title = {Multi-omics survival prediction under modality incompleteness via dynamic modality graphs and product-of-experts fusion},
journal = {Briefings in Bioinformatics},
year = {2026},
volume = {27},
number = {6},
doi = {10.1093/bib/bbag596}
}
EUVIMED