jax-unirep

A performant reimplementation of the UniRep protein featurization model in JAX.

The UniRep model was developed in George Church's lab, see the original publication here (bioRxiv) or here (Nature Methods), as well as the repository containing the original model.

This repo is a self-contained version of the UniRep model (so far only the 1900 hidden-unit mLSTM), with additional utility APIs that support protein engineering workflows.

Installation

Ensure that your compute environment allows you to run JAX code. (A modern Linux or macOS with a GLIBC>=2.23 is probably necessary.)

Then, you can install from PyPI:

pip install jax-unirep

If you want the latest and greatest, you can install directly from source:

pip install git+https://github.com/ElArkk/jax-unirep.git

More Details

Documentation on how to use jax-unirep is available here.

We highly encourage community contributions to this project! For more information, see the section in the docs on our contributing guidelines.

To read more about how we reimplemented the model in JAX, check out our preprint on bioarxiv.

License

All the model weights are licensed under the terms of Creative Commons Attribution-NonCommercial 4.0 International License. To view a copy of this license, visit here) or send a letter to Creative Commons, PO Box 1866, Mountain View, CA 94042, USA.

Otherwise the code in this repository is licensed under the terms of GPL v3.