Welcome to stko’s documentation!

GitHub: https://github.com/JelfsMaterialsGroup/stko

Tip

⭐ Star us on GitHub! ⭐

Overview

stko is a Python library for performing optimizations and calculations on complex molecules built using stk. In the case of optimizations, a clone of stk.Molecule is returned. For calculators, a Results class are used to calculate and extract properties of an stk.Molecule.

Installation

stko can be installed directly with pip:

pip install stko

Some optional dependencies are only available through conda:

# for xtb
mamba install xtb
# for openbabel, assuming you are not using Python >= 3.13!
mamba install openbabel

With OpenMM

To get stko and use OpenMM, we had some installation issues. The current solution is to first, in a new environment, install the OpenMM requirements:

mamba install -c conda-forge openff-toolkit

Then install stko with pip, but with the cuda variant to take advantage of GPU speed up (note that this is a heavy installation!).

pip install stko[cuda]

We also removed the default installation of espaloma_charge that provides the ML-based espaloma-am1bcc partial charges method. If users want this package, create a new environment and install their dependancies (if this fails, please check their instructions), then install stko:

mamba install -c conda-forge espaloma_charge openff-toolkit
pip install stko[cuda]

Developer Setup

Install just.
In a new virtual environment run:
```
$ just dev
```
Run code checks:
```
$ just check
```

Dependencies

The software packages we offer optimizers for are also depencies depending on the desired functions used. These are:

Examples

For every class (including Calculator, Optimizer), there are small examples of usage on the associated docs page. We have a page dedicated to analysing cage structures. There are also some examples for stko usage available here. These cover:

How To Contribute

If you have any questions or find problems with the code, please submit an issue.

If you wish to add your own code to this repository, please send us a Pull Request. Please maintain the testing and style that is used throughout `stko.

How To Cite

If you use stko please cite

https://github.com/JelfsMaterialsGroup/stko