This repo was mainly created for learning to create web apps using streamlit.
This tool predicts the molecular solubility of a compound given its SMILES string.
It also calculates shap values and plots some graphics to explain the models predictions.
It is possible to acess prediction from a Lineal Regression and a XGBoost models.
Clone the repository:
git clone https://github.com/jcorreia11/streamlit-solubility.git
Install the requirements:
pandas~=1.4.2
streamlit~=1.10.0
shap~=0.41.0
xgboost~=1.6.1
rdkit-pypi~=2022.3.3
streamlit_shap~=1.0.2
If you want to experiment with the solubility_models.ipynb notebook you also need:
scikit-learn~=1.0.1
matplotlib~=3.5.1
Run the following command in your terminal in the solubility_app.py directory:
streamlit run solubility_app.py
Paste your SMILES strings in the SMILES input box and explore!
- Heroku
- Streamlit Sharing