Flagship result
Quantum-computing-enhanced algorithm unveils potential KRAS inhibitors
A hybrid quantum–classical generative model for small-molecule design, aimed at KRAS, a protein long considered “undruggable” and a crucial target in cancer therapy. A quantum model trained on a 16-qubit IBM quantum computer was combined with classical generative methods to propose new molecules. Fifteen were synthesized and tested; two hold promise as inhibitors. The work shows that quantum computing can contribute to experimentally validated hits that compare favourably against classical models.
- Quantum–classical generative modelQuantum model trained on IBM hardware, combined with classical methods
- Design & selectNew KRAS-targeting molecules proposed and prioritised
- SynthesizeFifteen proposed molecules made in the laboratory
- Test against KRASExperimental assays: two molecules hold promise
In Nature Biotechnology: Cover story, December 2025 issue · Editors’ research picks: “2025: research in review”
In the news: U of T News · Temerty Faculty of Medicine · GEN
