Quantum-AI Hybrid Boosts Peptide Drug Discovery at DTU
Get the Health newsletter
Daily health & science — research, biotech, public health, the studies worth knowing. Free.
- DTU researchers showed a quantum-classical AI hybrid outperformed classical models at generating drug-discovery peptides, with the strongest gains where training data was rare.
- Timothy Patrick Jenkins and his team ran the project on weekends and pooled leftover money from other grants because, he says, "most innovative science is too scary for foundations."
- The team used ORCA Computing's printer-sized quantum computer to generate novel peptides — short amino acid chains that bind to specific proteins, a key step in vaccine development.
- Jonathan Funk, a DTU PhD student, cautioned that quantum can't yet encode normal-sized antibodies, limiting the current model to smaller peptide chains.
- ORCA CEO Richard Murray said the study is novel in showing a near-term commercial quantum application; ORCA also partners with BP on chemistry and Toyota on design efficiency.
- Jenkins sees the workflow as especially valuable for neglected diseases that receive little research funding — and is exploring quantum-enhanced AI for synthetic snakebite antidotes.
- A core challenge for the team is the lack of genetic data diversity — most medical research focuses on Western populations, making it hard to develop peptides for groups in Asia and Africa.
Why it matters: Jenkins's Novo Nordisk Foundation–funded team showed quantum-AI hybrids gain the most ground where training data is sparse — directly tackling drug discovery's data scarcity problem for underrepresented populations. ORCA CEO Murray frames it as one of the first clear near-term commercial uses of quantum, a field industrial firms still call "hazy and far away."
