Tangram2 is a computational framework for learning cell–cell communication directly from single-cell and spatial transcriptomics data. Tangram2 is organized into three modules:
Tangram2-mapping Aligns scRNA-seq profiles to spatial transcriptomics (SRT) by learning a probabilistic mapping from single cells to spots. This enables identification of co-localized cell populations while retaining transcriptome-wide coverage, and includes an integrated mapping mode to robustly handle both paired and unpaired scRNA/SRT datasets.
Tangram2-CCC Builds on the mapping output to infer cell–cell communication effects. It fits a linear model that decomposes gene expression into intrinsic cell-type programs and interaction-driven components, yielding an “interaction tensor” of gene-level coefficients that quantify how each sender cell type modulates each receiver cell type’s genes. This allows systematic identification of interaction-induced transcriptional programs across tissues.
Tangram2-evalkit Provides a synthetic data generation and benchmarking framework. Starting from real scRNA-seq with cell-type labels, it simulates Visium-like spot grids with configurable cell-type compositions, spatial structure, noise levels, and injected intercellular interactions. These datasets are then used to benchmark both mapping and CCC performance under controlled conditions.
To install tangram2, you can use the following command:
# Clone the repository
git clone https://github.com/Genentech/tangram2.git
# Change to the tangram2 directory
cd tangram2
# Create the environment using conda/mamba/micromamba
# Substitute {x} with your environment (eg macosx, linux, windows)
micromamba env create --file envs/{x}_environment.yml
# Activate the environment
micromamba activate tangram2
# Install the package
pip install .[cuda]For the environment, the X prefix represents the OS you are using, e.g., linux, macos, or windows.
Setting up the conda environment and completing the installation typically takes less than 10 minutes on a desktop computer.
The Tangram2 paper is published in Inferring cellular communication through mapping cells in space using Tangram 2 at BioRxiv(DOI: https://doi.org/10.1101/2025.09.28.679077). You can find all the notebooks used to generate the figures in the paper in the analysis directory. The notebooks are organized by figure number, and each notebook contains the code used to generate the corresponding figure in the paper.
Please make sure to use pre-commit when committing your changes. You can install pre-commit using the following command:
pip install pre-commitThen, run the following command to install the pre-commit hooks:
pre-commit installThis will ensure that your code follows the project's coding standards and style guidelines before you commit your changes.