MoESurv: a zero-sample and transferable survival prediction framework for rare cancers using mixture of experts
Bioinformatics · 2026
DOI: 10.1093/bioinformatics/btag549 ↗Kurz gefasst
Fachartikel in der Zeitschrift Bioinformatics, 2026. Verfasst von Shuping Fang und 9 weiteren Autor:innen. Beteiligte Einrichtung: Jiangsu Key Laboratory of Druggability of Biopharmaceuticals and State Key Laboratory of Natural Medicines, School of Life Science and Technology, Ch…. Hauptthemen laut MEDLINE: Neoplasms, Deep Learning, Computational Biology. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 43 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: the National Natural Science Foundation of China, Key R&D Program of Xinjiang Uygur Autonomous Region, Fundamental Research Funds for the Central Universities.
Abstract (Originalsprache)
Motivation Accurate survival prediction is crucial for personalized cancer treatment but remains challenging for rare cancers due to limited data. Most deep learning models require large training datasets, which are unavailable for rare cancer types, creating a significant clini-cal bottleneck. Results We propose MoESurv, a zero-sample survival prediction framework that leverages a mix-ture-of-experts architecture to extract generalizable prognostic patterns from pan-cancer data. The model integrates shared experts, cancer-specific experts, and routing experts within an autoencoder to disentangle common and type-specific survival features. Evaluated on seven rare TCGA cancer types, MoESurv achieved state-of-the-art performance, improving the average C-index by 4 percentage points over the best baseline. Further-more, external validation across diverse populations and independent cohorts—including a Chinese glioma cohort (CGGA mRNAseq_693, C-index = 0.7433), a rare GBM IDH-mutant subtype (C-index = 0.8064), and the pan-cancer PCAWG cohort (C-index = 0.7090)—demonstrated that MoESurv possesses the most robust predictive per-formance, highlighting its generalizability. MoESurv also effectively stratified high- and low-risk patient groups and identified potential survival-associated genes, demonstrating both clinical utility and biological interpretability. Availability The code is freely available at https://github.com/HuaYC666/MoESurv and https://zenodo.org/records/20785500.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Jiangsu Key Laboratory of Druggability of Biopharmaceuticals and State Key Laboratory of Natural Medicines, School of Life Science and Technology, China Pharmaceutical University Nanjing, Nanjing, Jiangsu, P.R. China.
- Regulatory Science, Jiangsu Simcere Pharmaceutical Co Ltd, Nanjing, Jiangsu, P.R. China.
- State Key Laboratory of Neurology and Oncology Drug Development, Jiangsu Simcere Pharmaceutical Co Ltd, Nanjing, Jiangsu, P.R. China.
- School of Science, China Pharmaceutical University, Nanjing, Jiangsu 211198, P.R. China.
- School of Life Science and Technology, China Pharmaceutical University, Nanjing, Jiangsu 211198, P.R. China.
- Mudi Meng Honors College, China Pharmaceutical University, Nanjing, Jiangsu 211198, P.R. China.
ORCID-Profile: Yuhang Wang, Jingya Fang, Hong Tian
Förderung
- the National Natural Science Foundation of China
- Key R&D Program of Xinjiang Uygur Autonomous Region
- Fundamental Research Funds for the Central Universities
Zitieren
APA
Fang, S., Wang, Y., Zhou, M., Shao, Y., Jiang, Z., Guo, Y., Jiang, T., Wang, F., Fang, J., & Tian, H. (2026). MoESurv: a zero-sample and transferable survival prediction framework for rare cancers using mixture of experts. Bioinformatics. https://doi.org/10.1093/bioinformatics/btag549
Vancouver
Fang S, Wang Y, Zhou M, Shao Y, Jiang Z, Guo Y, et al. MoESurv: a zero-sample and transferable survival prediction framework for rare cancers using mixture of experts. Bioinformatics. 2026. doi:10.1093/bioinformatics/btag549
BibTeX
@article{Fang2026,
author = {Fang, Shuping and Wang, Yuhang and Zhou, Mengyan and Shao, Yanting and Jiang, Zhenghao and Guo, Yitong and Jiang, Tao and Wang, Feng and Fang, Jingya and Tian, Hong},
title = {MoESurv: a zero-sample and transferable survival prediction framework for rare cancers using mixture of experts},
journal = {Bioinformatics},
year = {2026},
doi = {10.1093/bioinformatics/btag549}
}
EUVIMED