Toronto’s OpenPep receives US$2.3 million to build open AI tools for peptide discovery
Published in Artificial Intelligence, Bioinformatics, Drugs development & discovery, Tools.
Published October 14, 2026.
The University of Toronto announced US$2.3 million in Google.org funding for OpenPep, a project to build an open database and AI models for peptide drug discovery. Joseph Brown leads the initiative with collaborators Keith Pardee, Philip Kim and ProteinQure’s Chris Ing.
The team plans to produce up to 500 disease-relevant proteins using cell-free systems, then test peptide libraries with mass spectrometry to measure binding. These experimental results will train models intended to predict peptide interactions with proteins that have not yet been tested.
Researchers can nominate targets through the project’s website. Areas of interest include rare and pediatric diseases, cancer, neurodegeneration and cardiovascular disease.
The announcement describes funded research and a resource under development, rather than a completed platform or a clinically validated medicine. Predicted binding will still require experimental validation; binding alone does not establish a peptide’s safety or therapeutic effectiveness.