SpNeigh

SpNeigh provides methods for neighborhood-aware analysis of spatial transcriptomics data, including boundary detection, spatial weighting, spatially informed differential expression, and spatial enrichment analysis.

Key Features:

• Boundary Detection: Identifies tissue boundaries from spatial transcriptomics data.
• Spatial Weighting: Supports centroid-based and boundary-based spatial weighting schemes.
• Spatially Informed Differential Expression: Uses spline-based models for differential expression analysis that incorporates spatial information.
• Spatial Enrichment Index (SEI): Quantifies gene enrichment near defined spatial features.
• Spatial Neighborhood Extraction: Extracts spatial neighborhoods for local tissue context analysis.
• Distance-Based Weighting: Implements distance-based weighting for spatial analysis.
• Gradient-Based Statistical Testing: Performs statistical testing based on spatial gradients.
• Region-Based Differential Expression: Supports differential expression analysis between defined regions.
• Smooth Spatial Modeling: Uses spline-based regression for smooth spatial modeling of gene expression.
• Data Compatibility: Works with Seurat objects, SpatialExperiment objects, and spatial data frames.

Scientific Applications:

• Tissue Interface Analysis: Reveals intermediate cell populations at tissue boundaries.
• Immune Microenvironment Characterization: Identifies differences in immune microenvironments across spatial contexts.
• Spatial Zonation Pattern Detection: Discovers spatially zonated gene expression patterns in tissues.
• Cross-Platform Spatial Analysis: Applicable to Xenium, MERFISH, Visium HD, and other high-resolution spatial transcriptomics platforms.
• Complex Tissue Dissection: Enables analysis of spatial gene expression dynamics in tissues such as mouse brain, human breast cancer, and human liver.

Methodology:

SpNeigh employs boundary detection, spatial neighborhood extraction, distance-based weighting, and gradient-based statistical testing. It implements region-based differential expression and smooth spatial modeling using spline-based regression, along with a spatial enrichment index for identifying genes enriched near defined spatial features.