Python package to measure the similarity between matched single cell and targeted spatial transcriptomics data

In a clean conda environment with pip installed, run in the terminal:

git clone https://github.com/theislab/txsim.git

Navigate to the folder:

cd txsim

And install using pip:

pip install -e .

To import txsim, install squidpy (and all of its dependencies) into your environment For full functionality, the following are required as well:

• alphashape
• descartes
• pciSeq
• cellpose

All of the functions in txsim are currently either in the metrics or preprocessing module. A list of functions is as follows:

• coexpression_similarity
• coexpression_similarity_celltype
• all_metrics

• Normalization: normalize_total, normalize_pearson_residuals, normalize_by_area
• Segmentation: segment_nuclei, segment_cellpose
• Assignment: basic_assign, run_pciSeq
• Count Generation: generate_adata, calculate_alpha_area