spVC for the detection and interpretation of spatial gene expression variation.

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Spatially resolved transcriptomics enables gene expression analysis within tissue contexts.
• Current methods for identifying spatially variable genes often overlook continuous patterns and covariate integration.

Purpose of the Study:

• To introduce spVC, a novel statistical method for spatial transcriptomics.
• To address limitations in existing methods for identifying spatially variable genes.
• To enhance the integration of spot-level covariates in spatial gene expression analysis.

Main Methods:

• Developed spVC, a statistical method utilizing a generalized Poisson model.
• Integrated constant and spatially varying effects of covariates.
• Applied simulation and real-world spatial transcriptomics data for validation.

Main Results:

• spVC accurately identifies spatially variable genes.
• The method effectively integrates spot-level covariates.
• spVC demonstrates versatility in spatial transcriptomics analysis.

Conclusions:

• spVC offers a robust approach for exploring gene expression variability in spatial contexts.
• The method enhances the interpretability of spatial transcriptomics data.
• spVC represents a significant advancement for analyzing gene expression heterogeneity.