https://github.com/hasanulmukit/smiles2dta-demo
A Streamlit app for predicting drug-target binding affinity using a trained CNN model. Input SMILES strings and protein sequences for fast and accurate predictions.
https://github.com/hasanulmukit/smiles2dta-demo
bioinformatics cnn deep-learning drug-design drug-discovery drug-target-affinity drug-target-interactions machine-learning protein-sequence smiles streamlit
Last synced: 6 months ago
JSON representation
A Streamlit app for predicting drug-target binding affinity using a trained CNN model. Input SMILES strings and protein sequences for fast and accurate predictions.
- Host: GitHub
- URL: https://github.com/hasanulmukit/smiles2dta-demo
- Owner: hasanulmukit
- Created: 2025-01-27T10:48:40.000Z (9 months ago)
- Default Branch: main
- Last Pushed: 2025-04-05T04:43:24.000Z (6 months ago)
- Last Synced: 2025-04-05T05:28:26.480Z (6 months ago)
- Topics: bioinformatics, cnn, deep-learning, drug-design, drug-discovery, drug-target-affinity, drug-target-interactions, machine-learning, protein-sequence, smiles, streamlit
- Language: Jupyter Notebook
- Homepage: https://smiles2dta-demo.streamlit.app/
- Size: 18.1 MB
- Stars: 3
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# SMILES2DTA(DTC)\_Demo
🔬 **SMILES2DTA(DTC)\_Demo** is a Streamlit web application that predicts drug-target binding affinity using a trained deep learning model. It processes drug SMILES strings and protein sequences as inputs and provides the predicted binding affinity.
---
## Related Publication
This project is based on the following published research paper:
**SMILES2DTA: a CNN-based approach for identifying drug candidates and predicting drug-target binding affinity**
[Hasanul Mukit, Sayeed Hossain, Mirza Milan Farabi, Mehrab Zaman Chowdhury, Ahmed Iqbal Pritom & Humayan Kabir Rana]
[Neural Computing & Applications by Springer], [2024].
Link: [https://doi.org/10.1007/s00521-024-10814-x]
Please check the publication for a detailed explanation of the model and methodology.
## Features
- Accepts drug SMILES strings and protein sequences as inputs.
- Predicts binding affinity using a trained CNN model.
- User-friendly interface with clean and modern design.
---
## How It Works
- The user enters the drug's SMILES string and the protein sequence.
- The app tokenizes the inputs, pads them to a fixed length, and passes them to the trained model.
- The predicted binding affinity is displayed in the app.
## Technologies Used
- Jupyter Notebook
- Python
- TensorFlow/Keras
- Streamlit
- Pickle
---
### Acknowledgments
- This dashboard was created as part of a research project to simplify and improve drug-target binding affinity prediction.