Chemprop
A machine learning package implementing message passing neural networks for predicting molecular and reaction properties.
What is Chemprop?
Chemprop is a Python machine learning package that implements message passing neural networks for predicting molecular and reaction properties. It transforms chemical structures into graph representations and uses directed message passing neural networks to learn patterns that correlate with various chemical properties. The library has been successfully applied in drug discovery, including the identification of novel antibiotics like Halicin.
Computational chemists, drug discovery researchers, and machine learning practitioners working with chemical data who need to predict molecular properties, toxicity, or biological activity from structural information.
Chemprop provides specialized, well-tested implementations of graph neural networks optimized for chemical data, with proven success in real-world applications like antibiotic discovery. Its modular design and extensive documentation make it a reliable choice over generic deep learning frameworks for molecular property prediction tasks.
Overview
Message Passing Neural Networks for Molecule Property Prediction
Use Cases
Best For
- Predicting ADMET (absorption, distribution, metabolism, excretion, toxicity) properties for drug candidates
Related Projects
Found a gem we're missing?
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
