ai
Hybrid Quantum-Classical System Generates Peptides That Bind Proteins Better Than Classical Model Alone
A Technical University of Denmark team paired a generative AI model with a small quantum computer from ORCA Computing. Laboratory tests showed the hybrid setup produced more successful peptides than classical methods alone, especially with limited training data. The work focused on peptides that bind to specific proteins in the body.
Jul 12, 6:00 AM·1m read1 source
Audio version
Tap play to generate a narrated version.
Developing·Limited corroboration so far. This page will refresh as more sources emerge.
A team at the Technical University of Denmark ran a generative AI model for predicting proteins together with a printer-sized quantum computer built by British startup ORCA Computing. Wired reported that the hybrid quantum-classical system generated novel peptides capable of binding to specific proteins in the body.
Laboratory synthesis and binding tests showed the hybrid model produced more successful peptides than its classical counterpart.
The strongest performance gains occurred where training data was rare. The project was led by DTU professor Timothy Patrick Jenkins. The team worked weekends and used unspent funds from other projects because most innovative science is too scary for foundations, Jenkins said.
Jenkins stated he was a huge quantum skeptic and that the team needed to really prove it to convince skeptics that our predictions connect to the real world. DTU PhD student Jonathan Funk stated quantum is still not very powerful, so the level of complexity that we could encode wasn’t a normal-sized antibody, which is what we usually work with.
ORCA Computing CEO Richard Murray stated the study shows a near-term commercial application for quantum computing.
ORCA Computing is also working with BP on chemistry projects and with Toyota on design-process efficiency. The DTU team plans to test the workflow with more cutting-edge models and larger proteins. Jenkins is examining the use of a quantum computer to enhance generative AI for designing synthetic antidotes for snakebite venom.
