SpatialFlow

SpatialFlow is an open-source workflow/pipeline for analyzing spatial transcriptomics data.
It explore gene expression in tissue with spatial information.

Spatial transcriptomics is a growing field in biology. This project includes all basic and advanced steps — from loading data to generating results that can be used in research papers.

---

This Pipeline includes

• Load and explore spatial transcriptomics data
• Do quality control (QC) and filtering
• Normalize and reduce dimensions (PCA, UMAP)
• Cluster spatial spots or cells
• Detect marker genes
• Map spatial domains
• Estimate cell types in mixed spots (deconvolution)
• Study cell-cell communication (ligand-receptor analysis)
• Run pathway analysis
• Visualize and save results

---

1. Clone the repository:

   git clone https://github.com/rezwan-lab/spatialflow.git
   cd spatialflow

2. Create a conda environment

conda create -n spatialflow python=3.10

3. Activate SpatialFlow

conda activate spatialflow

4.1 Install dependencies

pip install -r requirements-spatialflow.txt

or use mamba
requirements-mamba.txt

4.2 Install the package in development mode

pip install -e .

5.1 Load test data

import squidpy as sq

# Load the example Visium dataset
adata = sq.datasets.visium("V1_Human_Lymph_Node")

# Print basic information about the loaded dataset
print(f"AnnData object: {adata}")
print(f"Shape of data matrix: {adata.shape}")
print(f"Available layers: {list(adata.layers.keys())}")
print(f"Spatial coordinates available: {'spatial' in adata.obsm}")

5.2 Convert data

import squidpy as sq

# Load the example Visium dataset (human lymph node)
adata = sq.datasets.visium("V1_Human_Lymph_Node")

# Save the data for SpatialFlow
adata.write_h5ad("./V1_Human_Lymph_Node.h5ad")

6. Generate config file

python -m spatialflow.cli init-config --output config.yaml
python -m spatialflow.cli init-config config.yaml

7. Usage

python -m spatialflow.cli --help

8.1 Run SpatialFlow workflow mode

python examples/basic_workflow.py

#or

python examples/advanced_workflow.py

8.2 Run SpatialFlow Analysis

python -m spatialflow.cli run <.h5ad file path> --config config.yaml --output-dir spatialflow_output

8.3 Run SpatialFlow visualization

python -m spatialflow.cli visualize spatialflow_output/data/*.h5ad -o spatialflow_output/figures

Author

Dr. Rezwanuzzaman Laskar