Curie Brief
Turn on cookies to sign in
Signing in saves your progress to your Curie account. We can only do that with cookies on — turn them on to continue.
NewsA University of Toronto-led team has secured $2.3 million from Google.org to build OpenPep — an open-access database and AI model designed to predict which peptides bind to disease-relevant proteins. Peptides are a fast-growing drug class, but discovery has been painfully slow due to fragmented data. OpenPep aims to change that by centralizing experimental data and training AI to accelerate the path from target to therapy.
A team of researchers at the University of Toronto has landed $2.3 million in funding from the Google.org Impact Challenge: AI for Science initiative to build OpenPep — an open-access database and AI model designed to predict peptide-protein binding. Peptides, short chains of amino acids that sit between small molecules and proteins in size, are a rapidly growing drug class: 8% of newly approved drugs are peptides, yet discovery has remained sluggish due to scattered, siloed data.
The project brings together experts in synthetic biology, mass spectrometry, and machine learning. The team plans to generate data on up to 500 disease-relevant proteins, test large peptide libraries against them, and use that experimental output to train AI models capable of predicting binding for proteins not yet tested in the lab. Target nominations are being crowd-sourced from hospitals, pharma companies, and research institutes, with a focus on pediatric and rare diseases, cancer, neurodegeneration, and cardiovascular disease.
Why it matters: Much like AlphaFold transformed protein structure prediction, OpenPep could become a foundational resource for peptide therapeutics — giving drug-discovery teams worldwide a data-rich head start and potentially shortening the timeline from disease target to viable treatment.