Installation

MolSysMT supports Python 3.13 (recommended), 3.12, and 3.11. High-performance compute kernels ship precompiled natively within the package.

---

Official Installation

The official distribution channel for MolSysMT is maintained under the uibcdf Conda channel. We strongly recommend installing MolSysMT within a dedicated Conda/Mamba environment:

conda install -c uibcdf -c conda-forge molsysmt

Or using mamba:

mamba install -c uibcdf -c conda-forge molsysmt

---

Development Version from Source

To test the latest features or contribute to the library, you can install the development version directly from the source repository on GitHub.

1. Clone the Repository

git clone https://github.com/uibcdf/molsysmt.git
cd molsysmt

2. Create the Development Conda Environment

Use the provided environment specification file located at devtools/conda-envs/development_env.yaml to create an isolated environment with Python 3.13 and all development dependencies:

conda env create -n molsysmt-dev -f devtools/conda-envs/development_env.yaml

3. Activate the Environment and Install in Editable Mode

Activate the newly created environment and install MolSysMT in editable mode without pulling redundant dependencies:

conda activate molsysmt-dev
pip install --no-deps --editable .

---

Precompiled Native Kernels

MolSysMT’s high-performance analytical kernels (distances, RMSD, SASA, gyration radius, dihedrals) are precompiled in Rust and integrated directly into the binary extension module.