https://github.com/ds4sd/subgrapher
[J. Cheminform.] SubGrapher: Visual Fingerprinting of Chemical Structures
https://github.com/ds4sd/subgrapher
Last synced: 5 months ago
JSON representation
[J. Cheminform.] SubGrapher: Visual Fingerprinting of Chemical Structures
- Host: GitHub
- URL: https://github.com/ds4sd/subgrapher
- Owner: DS4SD
- License: mit
- Created: 2025-04-07T13:38:26.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2025-10-14T09:14:47.000Z (9 months ago)
- Last Synced: 2025-10-14T18:22:24.611Z (9 months ago)
- Language: Python
- Homepage: https://arxiv.org/abs/2504.19695
- Size: 583 KB
- Stars: 4
- Watchers: 1
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# SubGrapher-IBM
[](https://huggingface.co/datasets/ds4sd/SubGrapher-Datasets)
[](https://doi.org/10.48550/arXiv.2504.19695)
[](https://jcheminf.biomedcentral.com/articles/10.1186/s13321-025-01091-4)
This is the repository for [SubGrapher: Visual Fingerprinting of Chemical Structures](https://jcheminf.biomedcentral.com/articles/10.1186/s13321-025-01091-4). SubGrapher is a model to detect functional groups in molecule images and convert them into visual fingerprints.
### Citation
If you find this repository useful, please consider citing:
```
@article{Morin2025,
title = {{Subgrapher: visual fingerprinting of chemical structures}},
author = {Morin, Lucas and Meijer, Gerhard Ingmar and Weber, Val{\'e}ry and Van Gool, Luc and Staar, Peter W. J.},
year = 2025,
month = {Sep},
day = 29,
journal = {Journal of Cheminformatics},
volume = 17,
number = 1,
pages = 149,
doi = {10.1186/s13321-025-01091-4},
issn = {1758-2946},
url = {https://doi.org/10.1186/s13321-025-01091-4}
}
```
### Installation
Create a virtual environment.
```
python3.11 -m venv subgrapher-env
source subgrapher-env/bin/activate
```
Install SubGrapher.
```
pip install -e .
```
### Inference
#### Script
1. Place your input images in: `SubGrapher/data/images/default/`.
2. Run SubGrapher:
```
python3 subgrapher/scripts/run.py
```
3. Read predictions in: `SubGrapher/data/predictions/default/`.
Predictions are structured as follows:
```
{
"substructure": "Amine_(tertiary)", # Predicted substructure
"confidence": 1.0, # Model confidence
"bbox": [525.3, 273.7, 675.8, 403.7], # Predicted bounding box
"type": "functional-groups" # Substructure type (functional-groups or carbon-chains)
}
...
```
4. (Optional) Read visualizations in: `SubGrapher/data/visualization/default/`.
Visualizations are structured as follows:
```
SubGrapher/data/visualization/default/
├── US07320977-20080122-C00078_fingerprint.png # Visualization of the SVMF
├── US07320977-20080122-C00078_substructures.png # Visualization of predicted substructures
├── US07320977-20080122-C00078_substructures.txt # Labels of predicted substructures
...
```
### Model
The model weights are available on [Hugging Face](https://huggingface.co/ds4sd/SubGrapher). They are automatically downloaded when running the model's inference.
### Evaluation Dataset
The benchmarks used for the visual fingerprinting evaluation are available on [HuggingFace](https://huggingface.co/datasets/ds4sd/SubGrapher-Datasets/) (download size: 50 MB).
```
wget https://huggingface.co/datasets/ds4sd/SubGrapher-Datasets/resolve/main/benchmarks.zip
```
The benchmarks are structured as follows:
```
benchmarks/
├── adenosine/ # Benchmark (Adenosine, Camphor, Cholesterol, Limonene, or Pyridine.)
│ ├── 90.csv/ # SMILES
│ └── adenosine/
│ ├── images_2/ # Images
│ └── molfiles/ # MolFiles
...
```