A spatially resolved human glioblastoma atlas reveals distinct cellular and molecular patterns of anatomical niches
Glioblastoma (GBM) is a highly aggressive brain tumor characterized by extreme cellular heterogeneity and poor clinical outcomes. Although single-cell analyses have uncovered diverse tumor and stromal cell states, their spatial architecture within the tumor microenvironment is still insufficiently defined. Here we present a multi-modal, spatially resolved atlas of human GBM by integrating spatial transcriptomics from both fresh frozen and FFPE tissue, matched single cell transcriptomic and proteomic data, and subcellular resolution in situ profiling. Using a custom 348 gene panel enriched for vascular markers, we map previously unrecognized GBM-associated endothelial, pericyte, and fibroblast states, and uncover their niche-specific associations with malignant and immune compartments. Notably, we identified a distinct oligodendrocyte subset confined to core and perivascular regions, characterized by transcriptional signatures predictive of recurrence and adverse clinical outcome. Our publicly accessible atlas provides a high-resolution framework for investigating GBM spatial biology and offers a resource for identifying spatially restricted therapeutic vulnerabilities.
The manuscript can be found here.
-
Raw Data Raw data generated in this study will be made available upon request.
-
Spatial Transcriptomics & Interactive Browsers Explore our spatial transcriptome and single-cell datasets via the following portals:
- SNUH-GliomaAtlas (GBM Visium) – A spatially resolved glioblastoma atlas featuring detailed transcriptomic maps across tissue sections.
- UCSC Multi-Omic GBM Browser – A comprehensive portal for exploring integrated multi-omic glioblastoma datasets.
-
Code & Video Tutorials All analysis code and walkthrough tutorials are available on GitHub: 👉 TSKGA Repository
-
Spatial Datasets (Visium, Xenium, IHC; Our Cohort & Public Data)
Available at: Amaranth InsightsLogin Credentials
- Email:
amaranthinsights@gmail.com - Password:
password@123
Workspace & Project Details
- Tenant Name:
Spatial Glioblastoma Atlas- You can verify the tenant by clicking the workspace panel next to the Go to the viewer button on the top right.
- Project Name:
Glioma manuscript
- Email:
-
Additional Materials Further information and resources are available from the corresponding authors upon request.
Video showing exploration of scRNA-seq, CITE-seq, and Visium spatial transcriptome datasets
Video showing exploration of Xenium spatial transcriptome dataset
Tools used:
RCCv2: https://github.com/nameetas/RCCv2
Data availability:
All the datasets can be downloaded or explored here.
In case of any queries please contact nameeta.shah@gmail.com.