Installation

Basic Installation

Install from PyPI:

pip install kompot

Or via conda/mamba:

conda install -c bioconda kompot

Optional Dependencies

Diffusion Maps

For using the default diffusion map cell state representation with Palantir:

pip install palantir

Plotting

For additional plotting functionality with scanpy integration:

pip install kompot[plot]

All Optional Dependencies

To install all optional dependencies:

pip install kompot[all]

JAX Installation

Kompot depends on JAX for efficient computations. By default, the CPU version of JAX is installed, which is recommended for most users as it provides good performance without memory constraints.

For GPU Support

If you want to use GPU acceleration, you’ll need to install the appropriate JAX version for your CUDA version. See the JAX installation guide for detailed instructions.

Example for CUDA 12:

pip install --upgrade "jax[cuda12]"

Example for CUDA 11:

pip install --upgrade "jax[cuda11]"

Note

GPU acceleration can significantly speed up computations for large datasets, but requires careful memory management. The CPU version is sufficient for most analyses and is recommended for getting started.

Development Installation

To install kompot from source for development:

git clone https://github.com/settylab/kompot.git
cd kompot
pip install -e ".[dev]"

This installs kompot in editable mode with development dependencies including testing and documentation tools.