RDE-PPI - GitLite

KO | EN

gitlite — search

#python #react #deep-learning #nodejs #macos #windows #awesome #kubernetes #vue #pytorch #ios #mcp #openai #open-source

RDE-PPI

(★ 63)

:mountain: Rotamer Density Estimator is an Unsupervised Learner of the Effect of Mutations on Protein-Protein Interaction (ICLR 2023)

File Explorer

Download Latest Version (.zip)

- cover.png
- - data_train_pdbredo_chain.yml
  - data_val_pdbredo_chain.yml
- - 7FAE_RBD_Fv_mutation.yml
- - rde.yml
  - rde_ddg_skempi.yml
- .gitignore
- 7FAE_RBD_Fv.fasta
- 7FAE_RBD_Fv.pdb
- get_pdbredo.sh
- get_skempi_v2.sh
- pdbredo_clusters.txt
- pdbredo_splits.txt
- rdelinear_params.csv
- - __init__.py
  - pdbredo_chain.py
  - pointmut.py
  - skempi.py
- - __main__.py
  - calibrate.py
  - entropy.py
  - export_params.py
- - __init__.py
  - rde.py
  - rde_ddg.py
- - data.py
  - misc.py
  - skempi.py
  - train.py
- .gitignore
.gitignore
env.yml
LICENSE
pointmut_analysis_entropy.py
pointmut_analysis_rdenet.py
README.md
test_rde_linear_skempi.py
test_rde_network_skempi.py
train_rde.py
train_rde_network_skempi.py

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# RDE-PPI
<img src="./assets/cover.png" alt="cover" style="width:50%;" />

:mountain: Rotamer Density Estimator is an Unsupervised Learner of the Effect of Mutations on Protein-Protein Interaction (ICLR 2023)

[[Paper]](https://www.biorxiv.org/content/10.1101/2023.02.28.530137)

## Install

### Environment

```bash
conda env create -f env.yml -n RDE
conda activate RDE
```

The default PyTorch version is 1.12.1 and cudatoolkit version is 11.3. They can be changed in [`env.yml`](./env.yml).

### Datasets

| Dataset   | Download Script                                    |
| --------- | -------------------------------------------------- |
| [PDB-REDO](https://pdb-redo.eu/)  | [`data/get_pdbredo.sh`](./data/get_pdbredo.sh)     |
| [SKEMPI v2](https://life.bsc.es/pid/skempi2) | [`data/get_skempi_v2.sh`](./data/get_skempi_v2.sh) |

### Trained Weights

https://drive.google.com/drive/folders/1jS7VCLyLZoxDyYb092Y7QqxyOGUE8iUm?usp=share_link

## Usage

### Evaluate RDE-Network

```bash
python test_rde_network_skempi.py --ckpt ./trained_models/DDG_RDE_Network_30k.pt
```

### Evaluate RDE-Linear

```
python test_rde_linear_skempi.py --ckpt ./trained_models/RDE.pt
```

### Predict Mutational Effects for a SARS-CoV-2 Human Antibody and Other Protein Complexes

Using the black-box **RDE-Net** model

```bash
python pointmut_analysis_rdenet.py ./configs/inference/7FAE_RBD_Fv_mutation.yml
```

or using the entropy-based **RDE-Linear** model

```bash
python pointmut_analysis_entropy.py ./configs/inference/7FAE_RBD_Fv_mutation.yml
```

💡 The script can also predict mutational effects for other protein complexes. Please refer to the example configuration file  [`7FAE_RBD_Fv_mutation.yml`](./configs/inference/7FAE_RBD_Fv_mutation.yml) to create your own mutation instruction.

### Train RDE

```bash
python train.py ./configs/train/rde.yml
```

### Train RDE-Network (DDG)

```bash
python train_rde_network_skempi.py ./configs/train/rde_ddg_skempi.yml
```

### Train RDE-Linear (DDG)

First, generate un-calibrated entropy for each mutation entry in the SKEMPI dataset:
```bash
python -m rde.linear.entropy
```

Then, train the coefficients in the linear model for ddG prediction, using the whole SKEMPI dataset for training (`--num_folds 1`):
```bash
python -m rde.linear.calibrate --num_folds 1
```

Finally, export the coefficients to `data/rdelinear_params.csv`:
```bash
python -m rde.linear.export_params
```

With the exported coefficients, you can predict ddG values using the `pointmut_analysis_entropy.py` script.

## Reference

```bibtex
@inproceedings{luo2023rotamer,
  title={Rotamer Density Estimator is an Unsupervised Learner of the Effect of Mutations on Protein-Protein Interaction},
  author={Shitong Luo and Yufeng Su and Zuofan Wu and Chenpeng Su and Jian Peng and Jianzhu Ma},
  booktitle={The Eleventh International Conference on Learning Representations },
  year={2023},
  url={https://openreview.net/forum?id=_X9Yl1K2mD}
}
```