Multi-task adaptive deep sparse canonical correlation analysis for multi-omics cancer survival prediction
PLOS One · 13.04.2026 · 21 (4) · e0346274
DOI: 10.1371/journal.pone.0346274 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS One, online erschienen am 13.04.2026. Verfasst von Yan Wang und 3 weiteren Autor:innen. Beteiligte Einrichtung: College of Life Sciences, Xinyang Normal University, Xinyang, Henan, China. Hauptthemen laut MEDLINE: Neoplasms. Die Arbeit nennt 24 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Nanhu Scholars Program for Young Scholars of Xinyang Normal University.
Abstract (Originalsprache)
Integrating high-dimensional multi-omics data is essential for uncovering the coordinated molecular mechanisms underlying cancer progression and improving survival prediction. DNA methylation and mRNA expression represent two tightly coupled regulatory layers; however, many existing approaches either model them independently or rely on linear assumptions that fail to capture the nonlinear cross-omics structure. Here, we propose MT-ADSCCA, a multitask adaptive deep sparse canonical correlation analysis framework that jointly learns correlated latent representations, selects interpretable multi-omics biomarkers, and supports downstream survival modeling. MT-ADSCCA embeds sparse CCA into a nonlinear encoder architecture and uses uncertainty-guided adaptive weighting to stabilize multi-objective training. The selected features were subsequently modeled using a BiLSTM–Cox survival network with genes ordered by chromosomal coordinates to capture local genomic dependencies. We evaluated MT-ADSCCA using event-stratified nested 10-fold cross-validation across three TCGA cohorts: breast invasive carcinoma (BRCA), glioma (GBMLGG), and pan-kidney carcinoma (KIPAN), including 485, 563, and 652 matched multi-omics samples, respectively. MT-ADSCCA achieved the highest concordance indices across all cohorts, outperforming six feature-selection baselines (DA, WGCNA, lmQCM, CCA, OSCCA, DeepCorrSurv) and four survival-model baselines (LASSO-Cox, RSF, MTLSA, DeepSurv). Kaplan–Meier analyses further confirmed a clear separation between the predicted high- and low-risk groups. The selected canonical features were enriched in biologically coherent functional categories, supporting the interpretability of the learned patterns. Together, these results demonstrate that MT-ADSCCA provides a robust and interpretable framework for multi-omics integration and cancer prognosis prediction.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Substanzen
Autor:innen und Institutionen
- College of Life Sciences, Xinyang Normal University, Xinyang, Henan, China.
- Department of Electrical, Computer and Software Engineering, University of Auckland, Auckland, New Zealand.
- College of International Education, Xinyang Normal University, Xinyang, Henan, China.
ORCID-Profile: Yan Wang
Förderung
- Nanhu Scholars Program for Young Scholars of Xinyang Normal University
Zitieren
APA
Wang, Y., Zou, Z., Wu, Y., & Chen, J. (2026). Multi-task adaptive deep sparse canonical correlation analysis for multi-omics cancer survival prediction. PLOS One, 21(4), e0346274. https://doi.org/10.1371/journal.pone.0346274
Vancouver
Wang Y, Zou Z, Wu Y, Chen J. Multi-task adaptive deep sparse canonical correlation analysis for multi-omics cancer survival prediction. PLOS One. 2026;21(4):e0346274. doi:10.1371/journal.pone.0346274
BibTeX
@article{Wang2026,
author = {Wang, Yan and Zou, Zimo and Wu, Yuanyuan and Chen, Jing},
title = {Multi-task adaptive deep sparse canonical correlation analysis for multi-omics cancer survival prediction},
journal = {PLOS One},
year = {2026},
volume = {21},
number = {4},
pages = {e0346274},
doi = {10.1371/journal.pone.0346274}
}
EUVIMED