Scientists Use AI and Quantum Computing to Discover New Peptides
▼ Summary
– A hybrid quantum-classical system generated more successful peptides for binding proteins than classical AI alone, especially when training data was scarce.
– The Technical University of Denmark team used spare funding and weekend work to run the experiment, as conventional grants often avoid innovative science.
– The approach could accelerate development of personalized immunotherapies and vaccines, including for understudied populations with limited genetic data.
– Quantum computers remain too small for full-scale AI models, so the study used small peptides rather than normal-sized antibodies.
– The research provides a near-term example of quantum computing’s commercial usefulness, though the technology still faces skepticism and technical hurdles.
Scientists have demonstrated for the first time that a quantum computer can significantly enhance the accuracy and range of generative AI drug discovery models. Remarkably, this breakthrough was achieved using spare time and leftover funds from other projects.
A team from the Technical University of Denmark integrated their generative AI model for predicting proteins with a printer-sized quantum computer built by British startup ORCA Computing. This hybrid approach linked quantum machines with traditional processors to accelerate the AI. The researchers used this technique to generate novel peptides, short chains of amino acids designed to bind to specific proteins in the body, a critical step in vaccine development.
The team worked weekends and pooled unspent money from other projects. According to DTU professor Timothy Patrick Jenkins, who led the project, “Most innovative science is too scary for foundations.”
When the peptides were synthesized in the lab and tested for their ability to bind to target proteins, the quantum-enhanced model produced a higher success rate than its classical counterpart. The most dramatic improvements occurred where training data was scarce, a common bottleneck in biomedical research.
The researchers believe this machine could accelerate the development of personalized immunotherapies and vaccines, as well as improve drug efficacy for understudied populations. “We needed to really prove it to convince skeptics that our predictions connect to the real world,” Jenkins told WIRED.
Quantum computing remains a nascent field facing intense scrutiny due to technical challenges. Even Jenkins was initially skeptical. “I was a huge quantum skeptic,” he says with a laugh, believing any application to his work would be “decades away.”
His team typically uses big data and AI to discover proteins that could unlock new immunotherapies more cheaply and quickly, often funded by the Novo Nordisk Foundation. A persistent challenge has been the lack of diverse genetic data, as most medical research has focused on Western populations. This data gap makes it difficult to develop peptides effective for populations in Asia and Africa.
The team hypothesized that embedding a quantum computer into their workflow could generate a more diverse set of peptides, especially for targets with limited data. This idea came after learning that quantum machines had a similar effect in generating images.
However, the process won’t revolutionize research yet. Quantum computers are still too small to run full-scale, cutting-edge AI models, meaning better results could often be achieved on a classical computer. “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,” says DTU PhD student Jonathan Funk. Moreover, finding a binding peptide is just one step in vaccine development and wouldn’t alone yield successful drugs.
“I think it’s no surprise that lots of industrial companies think quantum is hazy and far away,” says ORCA Computing CEO Richard Murray, partly because the technology “has not ever had really clear near-term examples of usefulness.” He calls this study novel for showing a near-term commercial application for quantum. ORCA is also working with oil major BP on chemistry and carmaker Toyota on design efficiency.
The DTU team will now explore using the workflow with more advanced models and larger proteins. “We needed this as an easy way to validate that now we actually have a shot at moving the needle substantially,” says Jenkins, noting that generative AI workflows are especially valuable for neglected diseases that receive little research funding. He is also investigating using a quantum computer to enhance his generative AI method for designing synthetic antidotes for snakebite venom.
(Source: Wired)
