SORC

SORC provides an integrated spatial transcriptomics resource across 17 cancer types, comprising 722,899 spatial spots from 269 slices and an integrated reference of 334,379 single-cell RNA-seq (scRNA-seq) cells across 46 cell types to enable analysis of tumor microenvironment spatial patterns.

Key Features:

• Slice Annotation: Annotates each slice with localized cellular composition and spatial labels for contextualizing cell distributions across 269 slices.
• Identification of Spatially Variable Genes: Detects genes with spatially heterogeneous expression within tumor tissues to reveal region-specific transcriptional programs.
• Co-occurrence Analysis of Immune Cells and Tumor Cells: Quantifies spatial co-localization patterns between immune cell types and tumor cells to characterize immune–tumor interactions.
• Functional Analysis: Identifies genes and pathways associated with spatial expression patterns to link local transcriptional changes to biological functions in cancer.
• Cell-Cell Communications: Analyzes intercellular signaling networks among annotated cell types to map communication patterns within the tumor microenvironment.

Scientific Applications:

• Tumor microenvironment mapping: Characterizing spatial organization of cell types and cellular states across 17 cancer types using high-resolution spot and slice data.
• Spatial biomarker discovery: Identifying spatially restricted genes and pathways that may serve as biomarkers of oncogenesis, progression, or metastasis.
• Immune–tumor interaction studies: Investigating spatial co-occurrence and interaction patterns between immune cells and tumor cells to inform immune response analyses.
• Cellular communication inference: Inferring cell–cell signaling networks to understand intercellular influences on tumor biology.
• Cross-platform cell-type refinement: Refining cell-type assignment in spatial data by integrating reference scRNA-seq (334,379 cells, 46 cell types).

Methodology:

Reference scRNA-seq data are matched to spatial transcriptomics datasets (matched across most datasets) and integrated; the integration comprises 334,379 scRNA-seq cells spanning 46 cell types.