SIU

KO | EN

gitlite — search

#javascript #react #typescript #llm #golang #rust #deep-learning #linux #windows #kubernetes #vue #ios #mcp #openai

(★ 17)

SIU: A Million-Scale Structural Small Molecule-Protein Interaction Dataset for Unbiased Bioactivity Prediction

SIU Latest Version Download

Download Latest Version (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# SIU

This GitHub repository contains the dataset and relevant code for the paper:

**SIU: A Million-Scale Structural Small Molecule-Protein Interaction Dataset for Unbiased Bioactivity Prediction**

## download the dataset

dataset is available at [huggingface dataset repo](https://huggingface.co/datasets/bgao95/SIU)

Please download and unzip all files

it should contains:

### gnn_cnn_data.zip

The dataset used to train gnn and cnn-3d model.

### split_60.zip

The dataset used to train Uni-Mol or ProFSA model for the SIU 0.6 version. It contains train.lmdb, valid.lmdb and test.lmdb.

### split_90.zip

The dataset used to train Uni-Mol or ProFSA model for the SIU 0.9 version. It contains train.lmdb, valid.lmdb and test.lmdb.

### pretrain_weights.zip

Pretrained weights for Uni-Mol and ProFSA.

profsa.pt: weights for pretrained ProFSA model

mol_pre_no_h_220816.pt: weights for pretrained Uni-Mol molecular Encoder

pocket_pre_220816.pt: weights for pretrained Uni-Mol pocketr Encoder

### final_dic.pkl

Complete dataset file in pickle format

### result_structures.zip

contains all structure files for protein and docked small molecules.

## dataset format

### final_dic.pkl

each key is an UniProt ID

corresponding value is a list of dictionaries. Each dictionary is a data point and has following keys:

| Key      | Description     |
|----------------|----------------|
| atoms  | atom types in ligand  |
| coordinates |  list of different conformations of the ligand |
|  pocket_atoms| atom types in pocket |
|  pocket_coordinates | atom positions of the pocket |
|  source_data |  UniProt ID and PDB ID information |
|  label |  dictionary for assay types and assay values |
|  ik | InChI key of the ligand |
|  smi |  SMILES notation of the ligand |

### Other lmdb files

All training and testing data are in lmdb format, and have the same keys as shown above.

Note that for single task learning, the label is a float value instead of a dictionary.

## Read Dataset

In read_data.py, we provide a script to read from lmdb files and pickle files.

## Environment

Follow the environment setting in [Uni-Mol](https://github.com/dptech-corp/Uni-Mol) and [Atom3D](https://github.com/drorlab/atom3d)

Or use **siu.yaml**

## Train the model with GNN/CNN-3D

```bash
cd ./atom3d/examples/lba
```

for cnn3d

```bash
cd cnn3d 
```

for gnn

```bash
cd gnn 
```

start training
```bash
python train.py

```

Note that the data path in train.py should be changed to atom3d_data/split_60 or atom3d_data/split_90

All the parameters are in train.py. We train the model with one NVIDIA A100 GPU.

## Train the model with ProFSA

```bash
cd ./unimol_train_code

bash binding_affinity_unimol.sh or binding_affinity_profsa.sh

```

use the pretrained weights in pretrain_weights.zip

Note that the data_path should in the bash file should be pointed to dir split_60 or split_90 (0.6 version and 0.9 version respectively)

If you want to to Multi Task Learning, please let --num-heads equals to 5, else set it to 1 and point to the correct directory in split_60 or split_90 (ic50, ec50, ki, kd)

All the parameters are in bash scripts. We train the model with 4 NVIDIA A100 GPU.

SIU

SIU Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

SIU

SIU Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

Download README