NAGL

Latest release
Status
Community

A playground for applying graph convolutional networks to molecules, with a focus on learning continuous "atom-type" embeddings and from these classical molecule force field parameters.

Note: This project is still in development and liable to substantial API and other changes.

This framework is mostly based upon the End-to-End Differentiable Molecular Mechanics Force Field Construction preprint by Wang, Fass and Chodera.

NAGL is bound by a Code of Conduct.

Documentation

See our documentation for notes on installation, basic usage, theory, and examples!

Copyright

The NAGL source code is hosted at https://github.com/openforcefield/openff-nagl and is available under the MIT license (see the file LICENSE). Some parts inherit from code distributed under other licenses, as detailed in LICENSE-3RD-PARTY).

NAGL inherits from Simon Boothroyd's NAGL library at https://github.com/SimonBoothroyd/nagl.