The Governing Council of the University of Toronto

Published purpose

Peptides are an increasingly effective and important drug modality that can serve to combine the advantages of small molecule drugs and large proteins. Yet, their discovery faces limitations in terms of cost and expertise, reducing access to drug discovery processes. Embracing in silico methods holds promise for revolutionizing biologic therapeutics discovery. However, high-quality training sets are rare, and training data must be generated specifically for each new drug target interaction. The goal of this project is to create a self-driving lab (SDL)-based discovery platform for peptides by combining droplet microfluidics, cell-free protein expression, and machine learning (ML)-guided design. This project will use split optical proteins, which are composed of two parts, a peptide and larger protein fragment, as well as insulin, as model peptides and targets for high-throughput screening. This platform will shift peptide discovery from manual to automated, accelerating the development of next-gen tools and drugs with broad applications in research and medicine.