STRUDEL

Substrate Prediction using Graph Neural Networks

Prototype

Contributed by Saad Naseem

This is a minimal web app prototype built during STRUDEL AI workshop, that lets users input a chemical name or SMILES ID and instantly receive predictions from a pre-trained ML model. Inspired by the CATNIP interface, this tool bridges cheminformatics and machine learning to explore molecular data in a user-friendly and browser based way.

Questions? Ideas? Looking to stay up to date?

Get in touch at strudel@lbl.gov

Join our mailing list by emailing strudel-community+subscribe@lbl.gov

Visit us on GitHub and join the Discussion

STRUDEL is an open source project housed at the Berkeley Institute for Data Science (BIDS) at the University of California, Berkeley. The project is generously funded by the Alfred P. Sloan Foundation, Liz Vu & Josh Greenberg Program Officers, grants G-2022-19360, G-2023-21098, and G-2024-22557. STRUDEL partners include members of the Lawrence Berkeley National Lab Scientific Data (SciData) Division UX team, Superbloom Design, The Carpentries, and 2i2c.