Stereo-seq Spatial Transcriptomics Analysis Pipeline

This repository contains a comprehensive analysis pipeline for Stereo-seq spatial transcriptomics data, focusing on gut tissue characterization and cell type deconvolution.

Overview

The pipeline processes Stereo-seq data at 50μm bin resolution to perform spatial cell type mapping, clustering analysis, and tissue organization studies. The analysis combines single-cell reference data with spatial transcriptomics to understand cellular composition and spatial patterns in gut tissue.

Cell Type Spatial Distribution

Spatial distribution of cell types across gut tissue showing epithelial and immune cell organization

Key Analysis Modules

Data Preprocessing (preprocessing_stomics.py)

• Quality control metrics calculation
• Gene and cell filtering
• Normalization and scaling
• Spatial coordinate handling

Cell Type Deconvolution (deconvolution_cell2loc.py)

• Cell2location-based spatial deconvolution
• Integration with gut tissue reference atlas
• Cell type abundance estimation per spatial location
• Supports bin50 Stereo-seq data format

Clustering Analysis (clustering_analysis.py)

• Spatial-aware clustering algorithms
• Multiple resolution clustering
• Cluster validation metrics
• Spatial coherence analysis

Visualization and Reporting

• Interactive dashboards (stomics_dashboad.py)
• Automated report generation (comprehensive_report_generator.py)
• OMERO integration for image management

Key Features

Spatial Cell Type Mapping

The pipeline uses Cell2location to deconvolve spatial transcriptomics data, identifying:

• Epithelial cells (Enterocytes, Goblet cells, Paneth cells)
• Immune cells (T cells, B cells, Plasma cells, Myeloid cells)
• Stromal cells (Mesenchymal, Fibroblasts)
• Endothelial cells (Vascular components)
• Neuronal cells (Enteric nervous system)

Analysis Capabilities

• Quality Control: Comprehensive QC metrics and filtering
• Spatial Clustering: Identify tissue regions and structures
• Cell Type Deconvolution: Estimate cell type abundances
• Gradient Analysis: Detect spatial gradients and transitions
• ROI Analysis: Focus on specific tissue regions
• Interactive Visualization: Real-time data exploration

Data Requirements

Input Data Format

• Spatial data: H5AD format with bin50 resolution
• Reference data: Single-cell RNA-seq atlas with cell type annotations
• Spatial coordinates: Embedded in AnnData.obsm['spatial']

Installation

Conda Environment

conda env create -f environment.yml
conda activate cell2loc_env


# Activate environment
conda activate cell2loc_env

# Run preprocessing
python src/preprocessing_stomics.py

# Perform cell type deconvolution
python src/deconvolution_cell2loc.py

# Generate clustering analysis
python src/clustering_analysis.py

# Create comprehensive report
python src/comprehensive_report_generator.py