This project was bootstrapped with Create React App and implements a web platform designed for predicting the biological activity of organic molecules based on SMILES notation, using machine learning models trained from molecular docking simulations.
🔬 This platform was developed as part of the research article titled:
“Exploring the molecular potential of natural products against COVID-19 through Molecular Modeling and Machine Learning”,
submitted to the journal Next Research.
PREDACTORS offers an intuitive interface for researchers and chemists to predict whether a molecule is active or inactive against a selected biological target.
- Access the platform: https://predactors.vercel.app
- Enter the SMILES code of the molecule in the input field.
- Wait 50 seconds to 1 minute while the system loads and processes the input through pre-trained machine learning models.
- View the prediction: the molecule will be classified as
ATIVO(active) orINATIVO(inactive) based on its predicted interaction with the biological target.
💡 No docking software is needed — predictions are made using chemical descriptors and ML algorithms trained from docking data.
In the project directory, you can run:
Runs the app in development mode.
Open http://localhost:3000 to view it in your browser.
Launches the test runner in interactive watch mode.
Builds the app for production to the build folder.
eject, you can't go back!
Rafael Vieira
Professor of Chemistry at the Federal Institute of Education, Science and Technology of Rondônia – Ji-Paraná – Brazil
ORCID: https://orcid.org/0000-0001-9003-3209
Vitor Hugo Batista
Undergraduate student in Chemistry at the Federal Institute of Education, Science and Technology of Rondônia – Ji-Paraná – Brazil