Related Experiment Video
Updated: May 24, 2025
14:27
Identification of Disease-related Spatial Covariance Patterns using Neuroimaging Data
Published on: June 26, 2013
15.6K
SpaFun: Discovering Domain-specific Spatial Expression Patterns and New Disease-Relevant Genes using Functional
Xi Jiang1,2, Yanghong Guo3, Lei Guo1
1Quantitative Biomedical Research Center, Peter O'Donnell Jr. School of Public Health, The University of Texas Southwestern Medical Center, Dallas, Texas, U.S.A.
Biorxiv : the Preprint Server for Biology
|March 3, 2025
Summary
SpaFun, a new method for spatial transcriptomics, efficiently identifies domain-representative genes (DRGs) by analyzing spatial heterogeneity and gene co-expression. It surpasses existing tools in accuracy and power, aiding discovery of novel disease-related genes.
Area of Science:
- Genomics
- Computational Biology
- Bioinformatics
Background:
- Spatially resolved transcriptomics (SRT) enables gene expression analysis within tissue context.
- Existing spatially variable gene (SVG) detection methods face computational and statistical limitations with large datasets.
- Current methods struggle to capture spatial heterogeneity and gene co-expression patterns effectively.
Purpose of the Study:
- To introduce SpaFun, a novel, non-model-based method for efficient and powerful SVG detection in large-scale SRT data.
- To address limitations of existing methods in computational efficiency, statistical power, and capturing spatial complexities.
- To identify domain-representative genes (DRGs) that accurately reflect spatial heterogeneity and co-expression.
Main Methods:
- Development of SpaFun based on functional principal component analysis (fPCA).
- Application of SpaFun to three diverse SRT datasets.
- Comparative analysis of SpaFun against state-of-the-art algorithms (DESeq, edgeR, limma, SPARK, CSIDE).
Main Results:
- SpaFun demonstrated superior computational efficiency and statistical power compared to existing SVG detection methods.
- The method accurately identified genes representative of specific spatial domains, such as tumor, immune, and stroma regions.
- SpaFun successfully uncovered novel disease-relevant genes missed by conventional and specialized spatial omics algorithms.
Conclusions:
- SpaFun offers a significant advancement in analyzing large-scale SRT data for identifying domain-representative genes.
- The method's ability to account for spatial heterogeneity and co-expression enhances biological insights.
- SpaFun has the potential to reveal new molecular mechanisms and inform therapeutic strategies for improved patient outcomes.
