Spatial Omics Skills

Skills for spatial transcriptomics data analysis, mapping, and visualization.

Available Skills

Single-Cell to Spatial Mapping

Map scRNA-seq to spatial data using optimal transport (MOSCOT) for gene imputation and cell type transfer.

Skill file: single_cell_spatial_mapping.md

When to use:

• You have paired scRNA-seq and spatial transcriptomics data
• You want to impute genes not measured in the spatial modality
• You want to transfer cell type annotations to spatial coordinates

3D Spatial Data Visualization

Interactive 3D visualization and rotating GIF animations for spatial data with PyVista.

Skill file: visualize_3d_spatial.md

When to use:

• Your spatial data has 3D coordinates
• You want to visualize gene expression or cell types in 3D
• You want to create rotating GIF animations

Spatial 3D Slice Alignment (Spateo)

Align serial spatial transcriptomics sections into a 3D volume using Spateo morpho_align with pairwise rigid registration.

Skill file: spatial_3d_alignment.md

When to use:

• You have serial tissue sections that need 3D reconstruction
• You want morphology + expression-based slice registration
• You need rigid transformations between consecutive sections

Spatial Cell-Cell Interaction (Spateo LR)

Infer ligand-receptor interactions between spatially adjacent cell types using Spateo's two-group CCI analysis with permutation testing.

Skill file: spatial_cci.md

When to use:

• You want to find LR interactions constrained by spatial proximity
• You have imputed spatial data with mapped cell type labels
• You want to compare spatial vs non-spatial CCI results

Spatial Deconvolution (Cell2location / Tangram)

Estimate cell type composition at each spatial location using scRNA-seq reference data. Two-stage model training with Cell2location, or simpler Tangram alternative.

Skill file: spatial_deconvolution.md

When to use:

• You want to estimate cell type proportions in spatial data
• You have a scRNA-seq reference with cell type annotations
• You want to impute gene expression via deconvolution

Spatial Signal Boundary Analysis

Detect expression domain boundaries between spatially antagonistic signals (e.g., Cer1 restricting Nodal). Includes auto-boundary detection, distance-decay analysis, and comprehensive 6-panel visualization.

Skill file: spatial_boundary_analysis.md

When to use:

• You have two spatially opposing signals (inhibitor/target)
• You want to quantify spatial restriction of expression domains
• You need publication-quality boundary analysis figures

Serial H&E Image Registration (RoMa)

Align consecutive H&E histology images using deep dense feature matching (RoMa + DINOv2) with RANSAC rigid transform estimation and BFS global composition.

Skill file: he_image_registration.md

When to use:

• You have serial H&E sections that need global alignment
• You want to build a 3D coordinate frame from histology images
• You need to co-register spatial transcriptomics data with H&E