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 Video

Updated: Jul 17, 2026

Generating the Transcriptional Regulation View of Transcriptomic Features for Prediction Task and Dark Biomarker Detection on Small Datasets
03:37

Generating the Transcriptional Regulation View of Transcriptomic Features for Prediction Task and Dark Biomarker Detection on Small Datasets

Published on: March 1, 2024

ViMST: vision transformer-based dual modality multi-task graph contrastive network for spatial transcriptomics

Cheng Ding1, Qiaoming Liu2, Yuming Zhao3

  • 1School of Computer Science and Artificial Intelligence, Northeast Forestry University, Harbin, 150040, China.

BMC Biology
|July 15, 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

Synthetic ecology of coastal ecosystems.

Cell reports·2026
Same author

Autonomic Signature-Driven Anesthesia Depth Monitoring with Biomimetic Wearable ECG and Knowledge Graph-Augmented Deep Networks.

Sensors (Basel, Switzerland)·2026
Same author

scTACL: a multitask topology-aware contrastive learning approach for single-cell transcriptomics analysis.

Bioinformatics (Oxford, England)·2026
Same author

Identification and characterization of lncRNA-stemness-immune regulatory patterns.

Briefings in bioinformatics·2026
Same author

Uniportal robotic-assisted left upper lobectomy with pulmonary artery reconstruction for a central lung cancer: a step-by-step surgical technique.

Journal of visualized surgery·2026
Same author

The double-edged sword of tumor-targeted treatment: the influence of targeted therapy on the heart.

Translational cancer research·2026

ViMST, a novel framework, integrates gene expression, image features, and spatial coordinates to analyze tissue microenvironments. This approach enhances spatial transcriptomics analysis, outperforming existing methods in identifying spatial domains and denoising data.

Area of Science:

  • Computational Biology
  • Bioinformatics
  • Genomics

Background:

  • Spatial transcriptomics is key to understanding cellular heterogeneity.
  • Current methods lack non-redundant information extraction and simultaneous spatial resolution of gene expression.
  • Tissue microenvironment analysis requires integrated multi-modal data.

Purpose of the Study:

  • To develop a novel framework for spatial transcriptomics analysis.
  • To integrate gene expression, image features, and spatial coordinates for enhanced tissue microenvironment exploration.
  • To improve the extraction of non-redundant information and spatial resolution of gene expression profiles.

Main Methods:

  • A vision transformer-based dual-modality multi-task graph contrastive network (ViMST).
Keywords:
Dual modality graph contrastive networkGene imputationSpatial clusteringSpatial transcriptomics

Related Experiment Videos

Last Updated: Jul 17, 2026

Generating the Transcriptional Regulation View of Transcriptomic Features for Prediction Task and Dark Biomarker Detection on Small Datasets
03:37

Generating the Transcriptional Regulation View of Transcriptomic Features for Prediction Task and Dark Biomarker Detection on Small Datasets

Published on: March 1, 2024

  • Utilizes Vision Transformer (ViT) for feature extraction.
  • Employs dual masked Graph Convolutional Networks (GCNs) and a joint topology decoder for multi-modal integration and relationship modeling.

    • Main Results:

    • ViMST outperforms eight state-of-the-art methods across nine spatial transcriptomics datasets.
    • Achieved superior performance in spatial domain identification and data denoising.
    • Demonstrated robust performance in data visualization, trajectory inference, identification of spatially variable genes (SVGs), horizontal integration, cellular heterogeneity analysis, and EMT studies.

    Conclusions:

    • ViMST is a powerful and versatile multimodal framework for spatial transcriptomics.
    • Its robust performance across diverse datasets and tasks demonstrates broad applicability in deciphering tissue spatial organization.
    • Enables comprehensive characterization of spatial heterogeneity, offering new avenues for disease mechanism understanding, biomarker discovery, and therapeutic target identification.