EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Expression graph network framework for biomarker discovery

Yang Liu, Jason Huse, Kasthuri Kannan iD

Briefings in Bioinformatics · 2025

DOI: 10.1093/bib/bbaf559 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Briefings in Bioinformatics, 2025. Verfasst von Yang Liu und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Translational Molecular Pathology, University of Texas MD Anderson Cancer Center, 2130 W Holcombe Blvd, Texas 77030, United States. Hauptthemen laut MEDLINE: Biomarkers, Tumor, Neural Networks, Computer, Neoplasms, Computational Biology. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 50 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: MD Anderson Moonshot Program.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Biomarker discovery for complex diseases, such as cancer, hinges on uncovering molecular signatures that capture intricate, interconnected relationships within biological data—a challenge that traditional statistical and machine learning methods often fail to meet due to the complexity of high-dimensional gene expression profiles. To overcome this, we introduce the expression graph network framework (EGNF). This cutting-edge graph-based approach integrates graph neural networks with network-based feature engineering to enhance the predictive identification of biomarkers. EGNF constructs biologically informed networks by combining gene expression data and clinical attributes within a graph database, utilizing hierarchical clustering to generate dynamic, patient-specific representations of molecular interactions. Leveraging graph learning techniques, including graph convolutional networks and graph attention networks, our framework identifies statistically significant and biologically relevant gene modules for classification. Validated across three independent datasets consisting of contrasting tumor types and clinical scenarios, EGNF consistently outperforms traditional machine learning models, achieving superior classification accuracy and interpretability. Notably, it delivers perfect separation between normal and tumor samples while excelling in nuanced tasks such as classifying disease progression and predicting treatment outcomes. This scalable, interpretable, and robust framework provides a powerful tool for biomarker discovery, with wide-ranging applications in precision medicine and the elucidation of disease mechanisms across diverse clinical contexts.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Biomarkers, TumorNeural Networks, ComputerNeoplasmsComputational BiologyGene Regulatory NetworksHumansMachine LearningGene Expression ProfilingBiomarkers

Stichwörter der Autor:innen

EGNFbiomarkergraph neural network

Substanzen

Biomarkers, TumorBiomarkers

Autor:innen und Institutionen

  • Department of Translational Molecular Pathology, University of Texas MD Anderson Cancer Center, 2130 W Holcombe Blvd, Texas 77030, United States.
  • Department of Biostatistics and Data Science, University of Texas Health Science Center at Houston, 1200 Pressler Street, Texas 77030, United States.
  • Department of Pathology, University of Texas MD Anderson Cancer Center, 1515 Holcombe Blvd, Texas 77030, United States.

ORCID-Profile: Kasthuri Kannan

Förderung

  • MD Anderson Moonshot Program

Zitieren

APA

Liu, Y., Huse, J., & Kannan, K. (2025). Expression graph network framework for biomarker discovery. Briefings in Bioinformatics. https://doi.org/10.1093/bib/bbaf559

Vancouver

Liu Y, Huse J, Kannan K. Expression graph network framework for biomarker discovery. Briefings in Bioinformatics. 2025. doi:10.1093/bib/bbaf559

BibTeX
@article{Liu2025,
  author = {Liu, Yang and Huse, Jason and Kannan, Kasthuri},
  title = {Expression graph network framework for biomarker discovery},
  journal = {Briefings in Bioinformatics},
  year = {2025},
  doi = {10.1093/bib/bbaf559}
}
Kostenlos starten →