Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

MCFST: spatial domain identification method based on multi-view graph convolutional network and graph fusion network.

Zilong Zhang1, Hao Duan2, Xin Gao3,4,5

  • 1Computer, Electrical and Mathematical Sciences and Engineering (CEMSE), King Abdullah University of Science and Technology (KAUST), Thuwal 23955-6900, Kingdom of Saudi Arabia.

Bioinformatics (Oxford, England)
|July 3, 2026
PubMed
Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Cross-coupled synchronization control strategy for rebar binding robots based on impedance control.

ISA transactions·2026
Same author

Multi-Frame Temporal Integration for 3-D Shape Measurement of Freely Falling Small Objects Using a High-Speed Camera Array.

Sensors (Basel, Switzerland)·2026
Same author

X-Linked USP11 Drives Depression-Like Behaviors by Stabilizing CK2α and Disrupting Mitochondrial Function.

CNS neuroscience & therapeutics·2026
Same author

Emerging frontiers in glioma therapy: the evolving role of glioma vaccines.

Frontiers in immunology·2026
Same author

Hypoxia Upregulates RIPK1 via the HIF-1α-JAK-STAT Pathway Leading to Astrocyte Necroptosis to Promote Cavitation After Spinal Cord Injury.

CNS neuroscience & therapeutics·2026
Same author

Inhibition of the PDGFRβ/MMP-9 pathway attenuates CUMS-induced blood-brain barrier disruption, neuroinflammation, and depressive-like behaviors.

International immunopharmacology·2026

MCFST, a novel graph neural network framework, enhances spatial domain identification in spatial transcriptomics by integrating multi-view data. This method improves accuracy and robustness in recognizing tissue regions with distinct molecular signatures.

Area of Science:

  • Genomics
  • Bioinformatics
  • Computational Biology

Background:

  • Spatial transcriptomics advances disease mechanism and developmental biology research.
  • Spatial domain identification is crucial for understanding tissue function and pathology.
  • Existing methods struggle to integrate heterogeneous spatial data, impacting accuracy.

Purpose of the Study:

  • To develop a robust method for spatial domain identification.
  • To address limitations in integrating multi-view spatial transcriptomics data.
  • To improve the accuracy and robustness of recognizing distinct tissue regions.

Main Methods:

  • Proposed MCFST, a graph neural network framework.
  • Integrated multi-view graph convolution with a mutual information-guided fusion module.

Related Experiment Videos

  • Aligned representations from gene expression, spatial coordinates, and expression profiles.

    • Main Results:

    • MCFST demonstrated superior performance in spatial domain identification across simulated and real datasets.
    • The method showed robustness against varying sparsity and noise levels.
    • Detected spatially variable genes from MCFST domains exhibited clear spatial expression patterns.

    Conclusions:

    • MCFST effectively captures latent patterns and achieves robust domain recognition.
    • The framework accurately identifies spatial domains, enhancing tissue function and pathology studies.
    • MCFST offers a robust and efficient solution for spatial transcriptomics data analysis.