RFDesign: Protein hallucination and inpainting with RoseTTAFold

Jue Wang (juewang@post.harvard.edu)
Doug Tischer (dtischer@uw.edu)
Sidney Lisanza (lisanza@uw.edu)
David Juergens (davidcj@uw.edu)
Joe Watson (jwatson3@uw.edu)

This repository contains code for protein hallucination or inpainting, as described in our preprint. Code for postprocessing and analysis scripts included in scripts/.

License

All code and neural network weights are open source under the BSD license. See LICENSE.

Installation

Clone the repository:

    git clone https://github.com/RosettaCommons/RFDesign.git
    cd rfdesign

Create environment and install dependencies:

    cd envs
    conda env create -f SE3.yml

Download weights (this step can be skipped if you downloaded the Zenodo version of this repo):

    cd hallucination/weights/rf_Nov05
    wget http://files.ipd.uw.edu/pub/rfdesign/weights/BFF_last.pt

    cd inpainting/weights/
    wget http://files.ipd.uw.edu/pub/rfdesign/weights/BFF_mix_epoch25.pt

Run tests

    cd hallucination/tests/
    ./run_tests.sh

    cd inpainting/tests/
    ./run_tests.sh

Dependencies

If you want/need to configure your environment manually, here are the packages in our environment:

python 3.8
pytorch 1.10.1
cudatoolkit 11.3.1
numpy
scipy
requests
packaging
pytorch-geometric (installation instructions)
dgl (installation instructions)
lie_learn
icecream (for inpainting.py)

Usage

See READMEs in hallucination/ and inpainting/ subfolders.

References

J. Wang, S. Lisanza, D. Juergens, D. Tischer, et al. Deep learning methods for designing proteins scaffolding functional sites. bioRxiv (2021). link

M. Baek, et al., Accurate prediction of protein structures and interactions using a three-track neural network, Science (2021). link

An earlier version of our hallucination method can be found at the trdesign-motif repo and published at:

D. Tischer, S. Lisanza, J. Wang, R. Dong, I. Anishchenko, L. F. Milles, S. Ovchinnikov, D. Baker. Design of proteins presenting discontinuous functional sites using deep learning. (2020) bioRxiv link

Our work is based on previous hallucination methods for unconstrained protein generation and fixed-backbone sequence design (trDesign repo):

I Anishchenko, SJ Pellock, TM Chidyausiku, ..., S Ovchinnikov, D Baker. De novo protein design by deep network hallucination. (2021) Nature link

C Norn, B Wicky, D Juergens, S Liu, D Kim, B Koepnick, I Anishchenko, Foldit Players, D Baker, S Ovchinnikov. Protein sequence design by conformational landscape optimization. (2021) PNAS link