Related Experiment Video
Updated: Aug 5, 2026
14:02
Super-Resolution Imaging to Study Co-Localization of Proteins and Synaptic Markers in Primary Neurons
Published on: October 31, 2020
Multi-Sample and Multi-Group Spatial Colocalization Analysis Using PANORAMIC
Jacob Chang1, Almudena Espín Pérez1, Perla Molina1
1Department of Biomedical Data Science, Stanford University, CA, USA.
Bioinformatics (Oxford, England)
|August 1, 2026
Summary
Spatial omics analysis often ignores within-sample uncertainty, potentially distorting results. PANORAMIC quantifies and propagates this uncertainty, improving cell colocalization inference in complex diseases like cancer.
Area of Science:
- Computational Biology
- Bioinformatics
- Spatial Omics
Background:
- Spatial omics studies compare cell organization across samples, but often overlook within-sample uncertainty.
- Ignoring this uncertainty can distort cohort-level analyses, especially in heterogeneous patient groups.
Purpose of the Study:
- To develop a method that quantifies and propagates within-sample uncertainty in spatial omics data.
- To improve the accuracy of cell colocalization inference in cohort-level studies.
Main Methods:
- PANORAMIC framework uses edge-corrected neighborhood enrichment for local colocalization.
- Spatial bootstrapping quantifies within-sample uncertainty.
- Multilevel random-effects meta-analysis propagates uncertainty across samples and conditions.
Main Results:
- PANORAMIC accurately recovers within-sample uncertainty and between-sample heterogeneity in simulations.
- It identified stronger B- and T-cell colocalization in colorectal tumors with Crohn's-like reactions.
- Found tighter immune organization in tumors, revealing findings missed by standard methods.
Conclusions:
- Propagating within-sample spatial uncertainty enhances cohort-level inference in spatial omics.
- PANORAMIC improves the understanding of immune cell organization in cancer.
- The method is robust to data degradation and applicable to diverse spatial settings.
