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 12, 2026

Mining Spatial Transcriptomics Datasets using DeepSpaceDB
10:16

Mining Spatial Transcriptomics Datasets using DeepSpaceDB

Published on: September 5, 2025

STAG: Biologically guided spatial transcriptomics prediction via hypergraph learning.

Mingcheng Qu1, Yuchuan Zhao1, Guang Yang1

  • 1Faculty of Computing, Harbin Institute of Technology, Harbin, China.

Medical Image Analysis
|July 9, 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

M<sup>3</sup>Surv: Fusing Multi-slide and Multi-omics for Memory-augmented robust Survival prediction.

Medical image analysis·2025
Same author

TV-3DG: Mastering Text-to-3D Customized Generation with Visual Prompt.

IEEE transactions on pattern analysis and machine intelligence·2025
Same author

Mode Hypergraph Neural Network.

IEEE transactions on neural networks and learning systems·2025
Same author

Generating Hypergraph-Based High-Order Representations of Whole-Slide Histopathological Images for Survival Prediction.

IEEE transactions on pattern analysis and machine intelligence·2022
Same author

Cost-Sensitive Uncertainty Hypergraph Learning for Identification of Lymph Node Involvement With CT Imaging.

Frontiers in medicine·2022
Same author

Big-Hypergraph Factorization Neural Network for Survival Prediction From Whole Slide Image.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2022
This summary is machine-generated.

STAG predicts spatial transcriptomics (ST) gene expression from histology images, improving accuracy by modeling complex spatial relationships within and between tissue slices. This computational approach enhances gene expression profiling and has applications in cancer classification.

Area of Science:

  • Computational biology
  • Genomics
  • Bioinformatics

Background:

  • Spatial transcriptomics (ST) provides spatially resolved gene expression data but is limited by high costs and low throughput.
  • Current computational methods for predicting gene expression from histology images are often restricted to 2D slices and do not fully capture complex spatial dependencies.

Purpose of the Study:

  • To develop a novel computational framework, STAG, for accurate gene expression prediction from histology images.
  • To model both individual spot gene expression and the structured relationships among multiple spots within and across tissue slices.
  • To leverage gene semantic information for enhanced, coordinated gene-aware predictions.

Main Methods:

  • Proposed STAG, a dual-branch framework with a Query branch for individual spot prediction and a Neighbor branch for spatial context modeling.
Keywords:
Contrastive learningCross-modal alignmentHypergraph learningSpatial transcriptomicsWhole-slide images

More Related Videos

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

Related Experiment Videos

Last Updated: Jul 12, 2026

Mining Spatial Transcriptomics Datasets using DeepSpaceDB
10:16

Mining Spatial Transcriptomics Datasets using DeepSpaceDB

Published on: September 5, 2025

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

  • Utilized hypergraph learning in the Neighbor branch to capture higher-order spatial and molecular dependencies, enabling unified 2D and 3D modeling.
  • Incorporated gene semantic information via foundation model encoding for biologically guided gene interactions.

    • Main Results:

    • STAG achieved an average gain of 5.16% in PCC@250 across six datasets.
    • Demonstrated superior performance with the lowest RMSE and highest PCC@50 under highly variable gene selection across three datasets.
    • Validated the effectiveness of learned representations in pseudo-3D prediction and cancer classification tasks.

    Conclusions:

    • STAG offers a powerful computational approach to predict spatial transcriptomics data from histology images, overcoming limitations of existing methods.
    • The framework effectively models complex spatial and molecular dependencies, extending from 2D to 3D scenarios.
    • STAG's gene-aware predictions and demonstrated utility in downstream tasks highlight its potential to advance single-cell genomics and precision medicine.