Awesome

pmpnndiff

Fast Non-autoregressive Inverse Folding with Discrete Diffusion (NeurIPS MLSB 2023)

This repository offers an implementation of discrete diffusion for inverse protein folding. It has pre-trained models, training routines, and inference scripts, ensuring a streamlined experience for protein folding predictions and experiments.

To-do List:

• Provide code of designability metric for proper reproducibility.
• Configure PMPNN ARM sampling temperature correctly.

1. Installation

Create and Activate Conda Environment

Clone the repository, navigate to its root directory, and create a conda environment using the provided YAML file. Activate the environment as follows:

conda env create -f environment.yml
conda activate your-env-name

Install Package Dependencies

Within the activated environment and the root directory of the repository, execute:

pip install -e .

2. Inference

Running Inference

For discrete diffusion inference with purity sampling, run

python experiments/inference_diff.py --sampling_type purity_sample

Refer to configs/clean/inference_diff.yaml for a complete description of inference args.

Compute Designability Numbers using ESMFold

To compute designability numbers, run

python scripts/run_esmfold_csv.py --csv_path your-csv-path ...

passing in your CSV path generated by inference_diff.py.

3. Weights

We provide pretrained weights for ProteinMPNN trained on the CATH 4.2 dataset under the weights directory. Both ARM and Discrete Diffusion weights are available.

4. Training Models

To train an ARM model from scratch, run

python experiments/train_arm.py ...

To train a discrete diffusion model from scratch, run

python experiments/train_diff.py ...

Contact

Please reach out to johnyang@mit.edu with any questions or concerns.

License

This project is endorsed under the MIT License - refer to the LICENSE.md file for details.