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 Concept Videos

You might also read

Related Articles

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

Sort by
Same author

Isavuconazole for invasive mold disease in patients with hematological malignancies: a multicenter real-world study from China on efficacy, safety, and competing risks.

Antimicrobial agents and chemotherapy·2026
Same author

A multi-scale graph frequency network for structural and functional region analysis in spatial transcriptomics.

Functional & integrative genomics·2026
Same author

CausalTCC: causal temporal contrastive learning for automated Alzheimer's disease biomarker discovery with bio-electrical signals.

Journal of neural engineering·2026
Same author

STELLA: a spatial transcriptomics framework for microenvironment decoding using dynamic graph neural networks.

Science China. Life sciences·2026
Same author

Integrated volatile profiling and metabolomics reveal changes in volatile compounds  during low-temperature smoking of sea bass (Lateolabrax maculatus).

Journal of the science of food and agriculture·2026
Same author

Risk prediction of pneumocystis jirovecii pneumonia in kidney transplant recipients using a clinical nomogram.

Frontiers in cellular and infection microbiology·2026
Same journal

Non-transferrin-bound iron uptake by cardiomyocytes.

Journal of molecular medicine (Berlin, Germany)·2026
Same journal

Beyond the target: implications of Wnt pathway inhibitors on bone health.

Journal of molecular medicine (Berlin, Germany)·2026
Same journal

Proteomic analysis reveals divergent inflammatory mechanisms of COVID-associated Guillain-Barré syndrome.

Journal of molecular medicine (Berlin, Germany)·2026
Same journal

Integrative genetic and functional analysis of autosomal dominant hearing loss in 108 multigenerational families.

Journal of molecular medicine (Berlin, Germany)·2026
Same journal

Preclinical evaluation of cysteine protease-inhibitor aloxistatin (E64d) for heart failure therapy.

Journal of molecular medicine (Berlin, Germany)·2026
Same journal

Zinc-binding protein metallothionein 3 protects vascular smooth muscle cells from ferroptosis via blocking lysosomal degradation of GPX4.

Journal of molecular medicine (Berlin, Germany)·2026
See all related articles

Related Experiment Video

Updated: Sep 14, 2025

Author Spotlight: Integrating Organoid Models with Single-Cell and Spatial Transcriptomics Technologies
05:45

Author Spotlight: Integrating Organoid Models with Single-Cell and Spatial Transcriptomics Technologies

Published on: March 29, 2024

2.6K

GraphCellNet: A deep learning method for integrated single-cell and spatial transcriptomic analysis with applications

Ruoyan Dai1, Zhenghui Wang1, Zhiwei Zhang1

  • 1Academy of Artificial Intelligence, Beijing Institute of Petrochemical Technology, Beijing, 102617, China.

Journal of Molecular Medicine (Berlin, Germany)
|July 21, 2025
PubMed
Summary
This summary is machine-generated.

GraphCellNet, a novel deep learning model, enhances spatial transcriptomics by accurately mapping cell types and spatial domains. This method improves understanding of tissue organization and development, offering new insights for regenerative medicine.

Keywords:
Cell type compositionGraph neural networksKolmogorov-Arnold NetworkSingle-cell RNA sequencingSpatial transcriptomics

More Related Videos

Multiplexed Single Cell mRNA Sequencing Analysis of Mouse Embryonic Cells
08:30

Multiplexed Single Cell mRNA Sequencing Analysis of Mouse Embryonic Cells

Published on: January 7, 2020

13.4K
Droplet Barcoding-Based Single Cell Transcriptomics of Adult Mammalian Tissues
10:12

Droplet Barcoding-Based Single Cell Transcriptomics of Adult Mammalian Tissues

Published on: January 10, 2019

18.7K

Related Experiment Videos

Last Updated: Sep 14, 2025

Author Spotlight: Integrating Organoid Models with Single-Cell and Spatial Transcriptomics Technologies
05:45

Author Spotlight: Integrating Organoid Models with Single-Cell and Spatial Transcriptomics Technologies

Published on: March 29, 2024

2.6K
Multiplexed Single Cell mRNA Sequencing Analysis of Mouse Embryonic Cells
08:30

Multiplexed Single Cell mRNA Sequencing Analysis of Mouse Embryonic Cells

Published on: January 7, 2020

13.4K
Droplet Barcoding-Based Single Cell Transcriptomics of Adult Mammalian Tissues
10:12

Droplet Barcoding-Based Single Cell Transcriptomics of Adult Mammalian Tissues

Published on: January 10, 2019

18.7K

Area of Science:

  • Genomics and Bioinformatics
  • Systems Biology
  • Computational Biology

Background:

  • Spatial transcriptomics (ST) integrates gene expression with spatial information for tissue analysis.
  • Single-cell RNA sequencing (scRNA-seq) aids ST but faces accuracy challenges in cell type deconvolution.
  • Existing methods struggle with ambiguous cell boundaries and high tissue heterogeneity.

Purpose of the Study:

  • To develop GraphCellNet, a novel deep learning model for accurate cell type deconvolution and spatial domain identification in ST data.
  • To enhance the modeling of nonlinear gene expression relationships and contextual integration within tissues.
  • To improve the analytical precision of ST data by addressing challenges in cell boundary definition and heterogeneity.

Main Methods:

  • Proposed GraphCellNet, a model combining cell type deconvolution and spatial domain identification.
  • Incorporated the Kolmogorov-Arnold Network (KAN) layer for enhanced nonlinear feature representation.
  • Utilized a graph-based approach for spatial domain identification, leveraging spatial relationships of cell types.
  • Evaluated performance using metrics such as PCC, SSIM, RMSE, JSD, and ARI.

Main Results:

  • GraphCellNet demonstrated high accuracy in cell type deconvolution and spatial domain identification across various biological systems.
  • Identified spatial regions with high Trem2 expression linked to metabolic gene signatures in myocardial infarction.
  • Uncovered TWEEDLE gene dynamics during Drosophila development.
  • Detailed cell compositions and spatial organization during human heart development.

Conclusions:

  • GraphCellNet offers a powerful deep learning framework for analyzing spatial transcriptomics data.
  • The KAN layer enhances the modeling of complex gene expression patterns efficiently.
  • The graph-based domain identification improves accuracy by considering spatial context.
  • The framework provides valuable insights into tissue organization and development, with implications for regenerative medicine.