EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Early cancer detection via multi-omics cfDNA fragmentation using early-late fusion neural network with sample-modality evaluation

Libo Lu iD, Yunze Wang, Xionghui Zhou iD

Briefings in Bioinformatics · 2025

DOI: 10.1093/bib/bbaf599 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Briefings in Bioinformatics, 2025. Verfasst von Libo Lu und 2 weiteren Autor:innen. Beteiligte Einrichtung: Hubei Key Laboratory of Agricultural Bioinformatics, College of Informatics, Huazhong Agricultural University, No.1 Shizishan Street, Hongshan Distri…. Hauptthemen laut MEDLINE: Neural Networks, Computer, Early Detection of Cancer, Biomarkers, Tumor, Neoplasms. Die Arbeit wurde bisher 10-mal zitiert (Crossref) und nennt 66 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: Biological Breeding Major Projects.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Cell-free DNA (cfDNA) fragmentation patterns reflect epigenetic modifications and are promising biomarkers for early cancer detection. While integrating diverse fragmentomic signals can improve accuracy, high modality dimensionality, and limited samples challenge effective multimodal fusion. We present Early-Late fusion with Sample-Modality evaluation (ELSM), a two-stage neural network integrating 13 fragmentomic feature spaces with sample-wise modality evaluation to capture complementary signals. Across five datasets of 1994 samples from 10 cancer types, ELSM outperforms unimodal and advanced multimodal models for cancer detection and tissue-of-origin prediction, achieving an AUC of 0.972 for pan-cancer diagnosis and 0.922 in an independent gastric cancer cohort, with a median tissue-of-origin accuracy of 0.683. Analysis of key genomic regions identified by ELSM reveals robust interpretability aligned with known oncogenic pathways. ELSM provides a powerful and interpretable framework for integrative multi-omics analysis with strong potential for clinical translation in early cancer detection.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Neural Networks, ComputerEarly Detection of CancerBiomarkers, TumorNeoplasmsGenomicsDNA FragmentationHumansMultiomics

Stichwörter der Autor:innen

cancer analysiscell-free DNAcohort studymachine learningmultimodal data processing

Substanzen

Biomarkers, Tumor

Autor:innen und Institutionen

  • Hubei Key Laboratory of Agricultural Bioinformatics, College of Informatics, Huazhong Agricultural University, No.1 Shizishan Street, Hongshan District, Wuhan, Hubei 430070, People's Republic of China.
  • Key Laboratory of Smart Farming for Agricultural Animals, Ministry of Agriculture and Rural Affairs, Huazhong Agricultural University, No.1 Shizishan Street, Hongshan District, Wuhan, Hubei 430070, People's Republic of…
  • College of Informatics, Huazhong Agricultural University, No. 1 Shizishan Street, Hongshan District, Wuhan, Hubei 430070, People's Republic of China.

ORCID-Profile: Libo Lu, Xionghui Zhou

Förderung

  • Biological Breeding Major Projects

Zitieren

APA

Lu, L., Wang, Y., & Zhou, X. (2025). Early cancer detection via multi-omics cfDNA fragmentation using early-late fusion neural network with sample-modality evaluation. Briefings in Bioinformatics. https://doi.org/10.1093/bib/bbaf599

Vancouver

Lu L, Wang Y, Zhou X. Early cancer detection via multi-omics cfDNA fragmentation using early-late fusion neural network with sample-modality evaluation. Briefings in Bioinformatics. 2025. doi:10.1093/bib/bbaf599

BibTeX
@article{Lu2025,
  author = {Lu, Libo and Wang, Yunze and Zhou, Xionghui},
  title = {Early cancer detection via multi-omics cfDNA fragmentation using early-late fusion neural network with sample-modality evaluation},
  journal = {Briefings in Bioinformatics},
  year = {2025},
  doi = {10.1093/bib/bbaf599}
}
Kostenlos starten →