https://github.com/m4marvin/plb_jnu
Deep Learning Framework for Protein-Ligand Binding Affinity Prediction
https://github.com/m4marvin/plb_jnu
bioinformatics biology chemistry com computational-biology deep-learning ligand protein python pytorch
Last synced: 3 months ago
JSON representation
Deep Learning Framework for Protein-Ligand Binding Affinity Prediction
- Host: GitHub
- URL: https://github.com/m4marvin/plb_jnu
- Owner: M4Marvin
- License: mit
- Created: 2025-01-11T07:37:34.000Z (9 months ago)
- Default Branch: main
- Last Pushed: 2025-01-28T17:45:58.000Z (8 months ago)
- Last Synced: 2025-03-15T14:12:39.009Z (7 months ago)
- Topics: bioinformatics, biology, chemistry, com, computational-biology, deep-learning, ligand, protein, python, pytorch
- Language: Jupyter Notebook
- Homepage:
- Size: 45.8 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Deep Learning Framework for Protein-Ligand Binding Affinity Prediction
## Overview
This research project implements a novel deep learning architecture combining 3D Convolutional Neural Networks (CNNs) and Graph Convolutional Networks (GCN) to predict protein-ligand binding affinities. The model processes structural and chemical information through parallel networks before combining outputs for final prediction.
## Architecture
- Dual 3D-CNN branches processing voxelized molecular representations (48×48×48×19)
- Graph Convolutional Network (GCN) for molecular topology
- Multi-Layer Perceptron (MLP) for final affinity prediction
## Dataset
- Based on the PDBbind dataset
- Preprocessed molecular structures with computed charges (MOL2 format)
- Voxelized representations for CNN input
## Requirements
```python
torch>=1.9.0
numpy
pandas
h5py
biopandas
```
## Project Structure
``` bash
├── LICENSE
├── README.md
├── data
│ ├── pdb_bind
│ │ ├── refined-set # Refined Set Data
│ │ └── v2020-other-PL # General Set
│ └── sample_data
├── notebooks
└── src
├── models
├── preprocessing
└── training
```
## Setup
- to be added.
## Usage
- to ber added.
## License
MIT
## Contact
[Your Name]
[Your Institution]
[Your Email]