Protein Workshop

Documentation

This repository provides the code for the protein structure representation learning benchmark detailed in the paper Evaluating Representation Learning on the Protein Structure Universe (ICLR 2024).

In the benchmark, we implement numerous featurisation schemes, datasets for self-supervised pre-training and downstream evaluation, pre-training tasks, and auxiliary tasks.

The benchmark can be used as a working template for a protein representation learning research project, a library of drop-in components for use in your projects, or as a CLI tool for quickly running protein representation learning evaluation and pre-training configurations.

Processed datasets and pre-trained weights are made available. Downloading datasets is not required; upon first run all datasets will be downloaded and processed from their respective source.

Configuration files to run the experiments described in the manuscript are provided in the proteinworkshop/config/sweeps/ directory.

Contents

• Protein Workshop
  • Contents
  • Installation
    • From PyPI
    • Building from source
  • Tutorials
  • Quickstart
    • Downloading datasets
    • Training a model
    • Finetuning a model
    • Running a sweep/experiment
    • Embedding a dataset
    • Visualising a dataset's embeddings
    • Performing attribution of a pre-trained model
    • Verifying a config
    • Using proteinworkshop modules functionally
  • Models
    • Invariant Graph Encoders
    • Equivariant Graph Encoders
      • (Vector-type)
      • (Tensor-type)
  • Datasets
    • Structure-based Pre-training Corpuses
    • Supervised Datasets
  • Tasks
    • Self-Supervised Tasks
    • Generic Supervised Tasks
  • Featurisation Schemes
    • Invariant Node Features
    • Equivariant Node Features
    • Edge Construction
    • Invariant Edge Features
    • Equivariant Edge Features
  • For Developers
    • Dependency Management
    • Code Formatting
    • Documentation

Installation

Below, we outline how one may set up a virtual environment for proteinworkshop. Note that these installation instructions currently target Linux-like systems with NVIDIA CUDA support. Note that Windows and macOS are currently not officially supported.

From PyPI

proteinworkshop is available for install from PyPI. This enables training of specific configurations via the CLI or using individual components from the benchmark, such as datasets, featurisers, or transforms, as drop-ins to other projects. Make sure to install PyTorch (specifically version 2.1.2 or newer) using its official pip installation instructions, with CUDA support as desired.

# install `proteinworkshop` from PyPI
pip install proteinworkshop

# install PyTorch Geometric using the (now-installed) CLI
workshop install pyg

# set a custom data directory for file downloads; otherwise, all data will be downloaded to `site-packages`
export DATA_PATH="where/you/want/data/" # e.g., `export DATA_PATH="proteinworkshop/data"`

However, for full exploration we recommend cloning the repository and building from source.

Building from source

With a local virtual environment activated (e.g., one created with conda create -n proteinworkshop python=3.10):

1. Clone and install the project

   git clone https://github.com/a-r-j/ProteinWorkshop
   cd ProteinWorkshop
   pip install -e .

2. Install PyTorch (specifically version 2.1.2 or newer) using its official pip installation instructions, with CUDA support as desired

   # e.g., to install PyTorch with CUDA 11.8 support on Linux:
   pip install torch==2.1.2+cu118 torchvision==0.16.2+cu118 torchaudio==2.1.2+cu118 --index-url https://download.pytorch.org/whl/cu118

3. Then use the newly-installed proteinworkshop CLI to install PyTorch Geometric

   workshop install pyg

4. Configure paths in .env (optional, will override default paths if set). See .env.example for an example.

5. Download PDB data:

   python proteinworkshop/scripts/download_pdb_mmtf.py

Tutorials

We provide a five-part tutorial series of Jupyter notebooks to provide users with examples of how to use and extend proteinworkshop, as outlined below.

1. Training a new model
2. Customizing an existing dataset
3. Adding a new dataset
4. Adding a new model
5. Adding a new task

Quickstart

Downloading datasets

Datasets can either be built from the source structures or downloaded from Zenodo. Datasets will be built from source the first time a dataset is used in a run (or by calling the appropriate setup() method in the corresponding datamodule). We provide a CLI tool for downloading datasets:

workshop download <DATASET_NAME>
workshop download pdb
workshop download cath
workshop download afdb_rep_v4
# etc..

If you wish to build datasets from source, we recommend first downloading the entire PDB first (in MMTF format, c. 24 Gb) to reuse shared PDB data as much as possible:

workshop download pdb
# or
python proteinworkshop/scripts/download_pdb_mmtf.py

Training a model

Launching an experiment minimally requires specification of a dataset, structural encoder, and task (devices can be specified with trainer=cpu/gpu):

workshop train dataset=cath encoder=egnn task=inverse_folding trainer=cpu env.paths.data=where/you/want/data/
# or
python proteinworkshop/train.py dataset=cath encoder=egnn task=inverse_folding trainer=cpu # or trainer=gpu

This command uses the default configurations in configs/train.yaml, which can be overwritten by equivalently named options. For instance, you can use a different input featurisation using the features option, or set the display name of your experiment on wandb using the name option:

workshop train dataset=cath encoder=egnn task=inverse_folding features=ca_bb name=MY-EXPT-NAME trainer=cpu env.paths.data=where/you/want/data/
# or
python proteinworkshop/train.py dataset=cath encoder=egnn task=inverse_folding features=ca_bb name=MY-EXPT-NAME trainer=cpu # or trainer=gpu

Finetuning a model

Finetuning a model additionally requires specification of a checkpoint.

workshop finetune dataset=cath encoder=egnn task=inverse_folding ckpt_path=PATH/TO/CHECKPOINT trainer=cpu env.paths.data=where/you/want/data/
# or
python proteinworkshop/finetune.py dataset=cath encoder=egnn task=inverse_folding ckpt_path=PATH/TO/CHECKPOINT trainer=cpu # or trainer=gpu

Running a sweep/experiment

We can make use of the hydra wandb sweeper plugin to configure experiments as sweeps, allowing searches over hyperparameters, architectures, pre-training/auxiliary tasks and datasets.

See proteinworkshop/config/sweeps/ for examples.

1. Create the sweep with weights and biases

wandb sweep proteinworkshop/config/sweeps/my_new_sweep_config.yaml

2. Launch job workers

With wandb:

wandb agent mywandbgroup/proteinworkshop/2wwtt7oy --count 8

Or an example SLURM submission script:

#!/bin/bash
#SBATCH --nodes 1
#SBATCH --ntasks-per-node=1
#SBATCH --gres=gpu:1
#SBATCH --array=0-32

source ~/.bashrc
source $(conda info --base)/envs/proteinworkshop/bin/activate

wandb agent mywandbgroup/proteinworkshop/2wwtt7oy --count 1

Reproduce the sweeps performed in the manuscript:

# reproduce the baseline tasks sweep (i.e., those performed without pre-training each model)
wandb sweep proteinworkshop/config/sweeps/baseline_fold.yaml
wandb agent mywandbgroup/proteinworkshop/2awtt7oy --count 8
wandb sweep proteinworkshop/config/sweeps/baseline_ppi.yaml
wandb agent mywandbgroup/proteinworkshop/2bwtt7oy --count 8
wandb sweep proteinworkshop/config/sweeps/baseline_inverse_folding.yaml
wandb agent mywandbgroup/proteinworkshop/2cwtt7oy --count 8

# reproduce the model pre-training sweep
wandb sweep proteinworkshop/config/sweeps/pre_train.yaml
wandb agent mywandbgroup/proteinworkshop/2dwtt7oy --count 8

# reproduce the pre-trained tasks sweep (i.e., those performed after pre-training each model)
wandb sweep proteinworkshop/config/sweeps/pt_fold.yaml
wandb agent mywandbgroup/proteinworkshop/2ewtt7oy --count 8
wandb sweep proteinworkshop/config/sweeps/pt_ppi.yaml
wandb agent mywandbgroup/proteinworkshop/2fwtt7oy --count 8
wandb sweep proteinworkshop/config/sweeps/pt_inverse_folding.yaml
wandb agent mywandbgroup/proteinworkshop/2gwtt7oy --count 8

Embedding a dataset

We provide a utility in proteinworkshop/embed.py for embedding a dataset using a pre-trained model. To run it:

python proteinworkshop/embed.py ckpt_path=PATH/TO/CHECKPOINT collection_name=COLLECTION_NAME

See the embed section of proteinworkshop/config/embed.yaml for additional parameters.

Visualising pre-trained model embeddings for a given dataset

We provide a utility in proteinworkshop/visualise.py for visualising the UMAP embeddings of a pre-trained model for a given dataset. To run it:

python proteinworkshop/visualise.py ckpt_path=PATH/TO/CHECKPOINT plot_filepath=VISUALISATION/FILEPATH.png

See the visualise section of proteinworkshop/config/visualise.yaml for additional parameters.

Performing attribution of a pre-trained model

We provide a utility in proteinworkshop/explain.py for performing attribution of a pre-trained model using integrated gradients.

This will write PDB files for all the structures in a dataset for a supervised task with residue-level attributions in the b_factor column. To visualise the attributions, we recommend using the Protein Viewer VSCode extension and changing the 3D representation to colour by Uncertainty/Disorder.

To run the attribution:

python proteinworkshop/explain.py ckpt_path=PATH/TO/CHECKPOINT output_dir=ATTRIBUTION/DIRECTORY

See the explain section of proteinworkshop/config/explain.yaml for additional parameters.

Verifying a config

python proteinworkshop/validate_config.py dataset=cath features=full_atom task=inverse_folding

Using proteinworkshop modules functionally

One may use the modules (e.g., datasets, models, featurisers, and utilities) of proteinworkshop functionally by importing them directly. When installing this package using PyPi, this makes building on top of the assets of proteinworkshop straightforward and convenient.

For example, to use any datamodule available in proteinworkshop:

from proteinworkshop.datasets.cath import CATHDataModule

datamodule = CATHDataModule(path="data/cath/", pdb_dir="data/pdb/", format="mmtf", batch_size=32)
datamodule.download()

train_dl = datamodule.train_dataloader()

To use any model or featuriser available in proteinworkshop:

from proteinworkshop.models.graph_encoders.dimenetpp import DimeNetPPModel
from proteinworkshop.features.factory import ProteinFeaturiser
from proteinworkshop.datasets.utils import create_example_batch

model = DimeNetPPModel(hidden_channels=64, num_layers=3)
ca_featuriser = ProteinFeaturiser(
    representation="CA",
    scalar_node_features=["amino_acid_one_hot"],
    vector_node_features=[],
    edge_types=["knn_16"],
    scalar_edge_features=["edge_distance"],
    vector_edge_features=[],
)

example_batch = create_example_batch()
batch = ca_featuriser(example_batch)

model_outputs = model(example_batch)

Read the docs for a full list of modules available in proteinworkshop.

Models

Invariant Graph Encoders

Name	Source	Protein Specific
GearNet	Zhang et al.	✓
DimeNet++	Gasteiger et al.	✗
SchNet	Schütt et al.	✗
CDConv	Fan et al.	✓

Equivariant Graph Encoders

(Vector-type)

Name	Source	Protein Specific
GCPNet	Morehead et al.	✓
GVP-GNN	Jing et al.	✓
EGNN	Satorras et al.	✗

(Tensor-type)

Name	Source	Protein Specific
Tensor Field Network	Corso et al.	✓
Multi-ACE	Batatia et al.	✗

Sequence-based Encoders

Name	Source	Protein Specific
ESM2	Lin et al.	✓

Datasets

To download a (processed) dataset from Zenodo, you can run

workshop download <DATASET_NAME>

where <DATASET_NAME> is given the first column in the tables below.

Otherwise, simply starting a training run will download and process the data from source.

Structure-based Pre-training Corpuses

Pre-training corpuses (with the exception of pdb, cath, and astral) are provided in FoldComp database format. This format is highly compressed, resulting in very small disk space requirements despite the large size. pdb is provided as a collection of MMTF files, which are significantly smaller in size than conventional .pdb or .cif file.

Name	Description	Source	Size	Disk Size	License
astral	SCOPe domain structures	SCOPe/ASTRAL		1 - 2.2 Gb	Publicly available
afdb_rep_v4	Representative structures identified from the AlphaFold database by FoldSeek structural clustering	Barrio-Hernandez et al.	2.27M Chains	9.6 Gb	GPL-3.0
afdb_rep_dark_v4	Dark proteome structures identied by structural clustering of the AlphaFold database.	Barrio-Hernandez et al.	~800k	2.2 Gb	GPL-3.0
afdb_swissprot_v4	AlphaFold2 predictions for SwissProt/UniProtKB	Kim et al.	542k Chains	2.9 Gb	GPL-3.0
afdb_uniprot_v4	AlphaFold2 predictions for UniProt	Kim et al.	214M Chains	1 Tb	GPL-3.0 / CC-BY 4.0
cath	CATH 4.2 40% split by CATH topologies.	Ingraham et al.	~18k chains	4.3 Gb	CC-BY 4.0
esmatlas	ESMAtlas predictions (full)	Kim et al.		1 Tb	GPL-3.0 / CC-BY 4.0
esmatlas_v2023_02	ESMAtlas predictions (v2023_02 release)	Kim et al.		137 Gb	GPL-3.0 / CC-BY 4.0
highquality_clust30	ESMAtlas High Quality predictions	Kim et al.	37M Chains	114 Gb	GPL-3.0 / CC-BY 4.0
igfold_paired_oas	IGFold Predictions for Paired OAS	Ruffolo et al.	104,994 paired Ab chains		CC-BY 4.0
igfold_jaffe	IGFold predictions for Jaffe2022 data	Ruffolo et al.	1,340,180 paired Ab chains		CC-BY 4.0
pdb	Experimental structures deposited in the RCSB Protein Data Bank	wwPDB consortium	~800k Chains	23 Gb	CC0 1.0

Additionally, we provide several species-specific compilations (mostly reference species)

Name	Description	Source	Size
a_thaliana	Arabidopsis thaliana (thale cress) proteome	AlphaFold2
c_albicans	Candida albicans (a fungus) proteome	AlphaFold2
c_elegans	Caenorhabditis elegans (roundworm) proteome	AlphaFold2
d_discoideum	Dictyostelium discoideum (slime mold) proteome	AlphaFold2
d_melanogaster	Drosophila melanogaster (fruit fly) proteome	AlphaFold2
d_rerio	Danio rerio (zebrafish) proteome	AlphaFold2
e_coli	Escherichia coli (a bacteria) proteome	AlphaFold2
g_max	Glycine max (soy bean) proteome	AlphaFold2
h_sapiens	Homo sapiens (human) proteome	AlphaFold2
m_jannaschii	Methanocaldococcus jannaschii (an archaea) proteome	AlphaFold2
m_musculus	Mus musculus (mouse) proteome	AlphaFold2
o_sativa	Oryza sative (rice) proteome	AlphaFold2
r_norvegicus	Rattus norvegicus (brown rat) proteome	AlphaFold2
s_cerevisiae	Saccharomyces cerevisiae (brewer's yeast) proteome	AlphaFold2
s_pombe	Schizosaccharomyces pombe (a fungus) proteome	AlphaFold2
z_mays	Zea mays (corn) proteome	AlphaFold2

Supervised Datasets

Name	Description	Source	License
antibody_developability	Antibody developability prediction	Chen et al.	CC-BY 3.0
atom3d_msp	Mutation stability prediction	Townshend et al.	MIT
atom3d_ppi	Protein-protein interaction prediction	Townshend et al.	MIT
atom3d_psr	Protein structure ranking	Townshend et al.	MIT
atom3d_res	Residue identity prediction	Townshend et al.	MIT
ccpdb_ligands	Ligand binding residue prediction	Agrawal et al.	Publicly Available
ccpdb_metal	Metal ion binding residue prediction	Agrawal et al.	Publicly Available
ccpdb_nucleic	Nucleic acid binding residue prediction	Agrawal et al.	Publicly Available
ccpdb_nucleotides	Nucleotide binding residue prediction	Agrawal et al.	Publicly Available
deep_sea_proteins	Gene Ontology prediction (Biological Process)	Sieg et al.	Public domain
go-bp	Gene Ontology prediction (Biological Process)	Gligorijevic et al	CC-BY 4.0
go-cc	Gene Ontology (Cellular Component)	Gligorijevic et al	CC-BY 4.0
go-mf	Gene Ontology (Molecular Function)	Gligorijevic et al	CC-BY 4.0
ec_reaction	Enzyme Commission (EC) Number Prediction	Hermosilla et al.	MIT
fold_fold	Fold prediction, split at the fold level	Hou et al.	CC-BY 4.0
fold_family	Fold prediction, split at the family level	Hou et al.	CC-BY 4.0
fold_superfamily	Fold prediction, split at the superfamily level	Hou et al.	CC-BY 4.0
masif_site	Protein-protein interaction site prediction	Gainza et al.	Apache 2.0
metal_3d	Zinc Binding Site Prediction	Duerr et al.	MIT
ptm	Post Translational Modification Side Prediction	Yan et al.	CC-BY 4.0

Tasks

Self-Supervised Tasks

Name	Description	Source
inverse_folding	Predict amino acid sequence given structure
residue_prediction	Masked residue type prediction
distance_prediction	Masked edge distance prediction	Zhang et al.
angle_prediction	Masked triplet angle prediction	Zhang et al.
dihedral_angle_prediction	Masked quadruplet dihedral prediction	Zhang et al.
multiview_contrast	Contrastive learning with multiple crops and InfoNCE loss	Zhang et al.
structural_denoising	Denoising of atomic coordinates with SE(3) decoders

Generic Supervised Tasks

Generic supervised tasks can be applied broadly across datasets. The labels are directly extracted from the PDB structures.

These are likely to be most frequently used with the pdb dataset class which wraps the PDB Dataset curator from Graphein.

Name	Description	Requires
binding_site_prediction	Predict ligand binding residues	HETATM ligands (for training)
ppi_site_prediction	Predict protein binding residues	graph_y attribute in data objects specifying the desired chain to select interactions for (for training)

Featurisation Schemes

Part of the goal of the proteinworkshop benchmark is to investigate the impact of the degree to which increasing granularity of structural detail affects performance. To achieve this, we provide several featurisation schemes for protein structures.

Invariant Node Features

N.B. All angular features are provided in [sin, cos] transformed form. E.g.: $\textrm{dihedrals} = [sin(\phi), cos(\phi), sin(\psi), cos(\psi), sin(\omega), \cos(\omega)]$, hence their dimensionality will be double the number of angles.

Name	Description	Dimensionality
residue_type	One-hot encoding of amino acid type	21
positional_encoding	Transformer-like positional encoding of sequence position	16
alpha	Virtual torsion angle defined by four $C_\alpha$ atoms of residues $I_{-1}, I, I_{+1}, I_{+2}$	2
kappa	Virtual bond angle (bend angle) defined by the three $C_\alpha$ atoms of residues $I_{-2}, I, I_{+2}$	2
dihedrals	Backbone dihedral angles $(\phi, \psi, \omega)$	6
sidechain_torsions	Sidechain torsion angles $(\chi_{1-4})$	8

Equivariant Node Features

Name	Description	Dimensionality
orientation	Forward and backward node orientation vectors (unit-normalized)	2

Edge Construction

We predominanty support two types of edges: $k$-NN and $\epsilon$ edges.

Edge types can be specified as follows:

python proteinworkshop/train.py ... features.edge_types=[knn_16, knn_32, eps_16]

Where the suffix after knn or eps specifies $k$ (number of neighbours) or $\epsilon$ (distance threshold in angstroms).

Invariant Edge Features

Name	Description	Dimensionality
edge_distance	Euclidean distance between source and target nodes	1
node_features	Concatenated scalar node features of the source and target nodes	Number of scalar node features $\times 2$
edge_type	Type annotation for each edge	1
sequence_distance	Sequence-based distance between source and target nodes	1
pos_emb	Structured Transformer-inspired positional embedding of $i - j$ for source node $i$ and target node $j$	16

Equivariant Edge Features

Name	Description	Dimensionality
edge_vectors	Edge directional vectors (unit-normalized)	1

For Developers

Dependency Management

We use poetry to manage the project's underlying dependencies and to push updates to the project's PyPI package. To make changes to the project's dependencies, follow the instructions below to (1) install poetry on your local machine; (2) customize the dependencies; or (3) (de)activate the project's virtual environment using poetry:

1. Install poetry for platform-agnostic dependency management using its installation instructions

   After installing poetry, to avoid potential keyring errors, disable its keyring usage by adding PYTHON_KEYRING_BACKEND=keyring.backends.null.Keyring to your shell's startup configuration and restarting your shell environment (e.g., echo 'export PYTHON_KEYRING_BACKEND=keyring.backends.null.Keyring' >> ~/.bashrc && source ~/.bashrc for a Bash shell environment and likewise for other shell environments).

2. Install, add, or upgrade project dependencies

     poetry install  # install the latest project dependencies
     # or
     poetry add XYZ  # add dependency `XYZ` to the project
     # or
     poetry show  # list all dependencies currently installed
     # or
     poetry lock  # standardize the (now-)installed dependencies

3. Activate the newly-created virtual environment following poetry's usage documentation

     # activate the environment on a `posix`-like (e.g., macOS or Linux) system
     source $(poetry env info --path)/bin/activate

     # activate the environment on a `Windows`-like system
     & ((poetry env info --path) + "\Scripts\activate.ps1")

     # if desired, deactivate the environment
     deactivate

Code Formatting

To keep with the code style for the proteinworkshop repository, using the following lines, please format your commits before opening a pull request:

# assuming you are located in the `ProteinWorkshop` top-level directory
isort .
autoflake -r --in-place --remove-unused-variables --remove-all-unused-imports --ignore-init-module-imports .
black --config=pyproject.toml .

Documentation

To build a local version of the project's Sphinx documentation web pages:

# assuming you are located in the `ProteinWorkshop` top-level directory
pip install -r docs/.docs.requirements # one-time only
rm -rf docs/build/ && sphinx-build docs/source/ docs/build/ # NOTE: errors can safely be ignored

Citing ProteinWorkshop

Please consider citing proteinworkshop if it proves useful in your work.

@inproceedings{
  jamasb2024evaluating,
  title={Evaluating Representation Learning on the Protein Structure Universe},
  author={Arian R. Jamasb, Alex Morehead, Chaitanya K. Joshi, Zuobai Zhang, Kieran Didi, Simon V. Mathis, Charles Harris, Jian Tang, Jianlin Cheng, Pietro Lio, Tom L. Blundell},
  booktitle={The Twelfth International Conference on Learning Representations},
  year={2024},
}