Related Experiment Video
Updated: Jul 16, 2026
09:19
Spatial Profiling of Protein and RNA Expression in Tissue: An Approach to Fine-Tune Virtual Microdissection
Published on: July 6, 2022
5.3K
Spatial Transcriptomics As Rasterized Image Tensors (STARIT) characterizes cell states with subcellular molecular
Dee Velazquez1,2, Caleb Hallinan1,2, Roujin An1,2
1Center for Computational Biology, Whiting School of Engineering, Johns Hopkins University, Baltimore, Maryland, USA.
Biorxiv : the Preprint Server for Biology
|January 7, 2026
Summary
STARIT converts spatial transcriptomics data into image tensors, enabling deep learning analysis. This method captures subcellular transcript localization to identify cell types and states missed by traditional gene counting.
Area of Science:
- Genomics
- Bioinformatics
- Computational Biology
Background:
- Imaging-based spatially resolved transcriptomics (imSRT) offers high-throughput, molecular-resolution spatial gene characterization within cells.
- Conventional imSRT analysis uses gene count matrices, overlooking subcellular transcript heterogeneity crucial for defining cell states.
Purpose of the Study:
- To develop a novel method, STARIT (Spatial Transcriptomics As Rasterized Image Tensors), for analyzing imSRT data.
- To leverage subcellular transcript localization for enhanced cell-type and cell-state identification.
Main Methods:
- STARIT converts imSRT data into image-based tensor representations.
- Integrates these tensors with deep learning computer vision models for downstream analysis.
- Validates performance using simulated and real imSRT datasets.
Main Results:
- STARIT successfully distinguishes transcriptionally distinct cell types and separates cell states based on subcellular transcript localization in simulated data.
- On real imSRT data, STARIT identified comparable cell types to conventional methods and revealed rotational variations.
- The method captures biological insights missed by traditional gene count matrices.
Conclusions:
- STARIT provides a standardized framework to encode subcellular molecular information from imSRT data.
- Enables deeper insights into cellular heterogeneity and improves identification of cell types and states.
- Facilitates advanced analysis of imSRT data by integrating spatial transcriptomics with deep learning.
Related Concept Videos
01:58Cell Specific Gene Expression
Multicellular organisms contain a variety of structurally and functionally distinct cell types, but the DNA in all the cells originated from the same parent cells. The differences in the cells can be attributed to the differential gene expression. Liver cells, whose functions include detoxification of blood, production of bile to metabolize fats, and synthesis of proteins essential for metabolism, must express a specific set of genes to perform their functions. Gene expression also varies with...
01:58Cell Specific Gene Expression
Multicellular organisms contain a variety of structurally and functionally distinct cell types, but the DNA in all the cells originated from the same parent cells. The differences in the cells can be attributed to the differential gene expression. Liver cells, whose functions include detoxification of blood, production of bile to metabolize fats, and synthesis of proteins essential for metabolism, must express a specific set of genes to perform their functions. Gene expression also varies with...