Quantum Computing and AI: Revolutionizing Drug Development
Scientists have succeeded demonstrated that a quantum computer can improve the accuracy and reach of drug development models using generative artificial intelligence. And they did it with their free time and the money they had left over from other projects.
A Leap Forward in Biotech
The team from the Technical University of Denmark ran their generative AI protein prediction model in conjunction with a printer-sized quantum computer from British startup ORCA Computing, which accelerated AI by linking quantum machines to conventional processors. The researchers used the hybrid technique to create novel peptides – short chains of amino acids – that can bind to specific proteins in the body. This is a crucial step in vaccine development.
Overcoming Financial and Temporal Constraints
The research team worked on weekends and collected unspent money from other projects because “the most innovative science is too scary for foundations,” said DTU professor Timothy Patrick Jenkins, who led the project.
Proving the Model’s Efficacy
Making the peptides in the lab and testing whether they would bind to the respective proteins showed that the model produced more successful peptides than its classical counterpart, with the strongest improvements occurring where training data was rare.
Implications for Personalized Medicine
The team believes the machine could accelerate the development of personalized immunotherapies and vaccines and improve the effectiveness of drugs in under-researched groups.
Overcoming Skepticism
“We really had to prove it to convince skeptics that our predictions were consistent with the real world,” Patrick Jenkins tells WIRED. Quantum computing remains a nascent field and faces intense scrutiny due to the technical challenges in building these machines and successfully applying them to solve problems.
Even Patrick Jenkins was initially hesitant to explore the technology: “I was a big quantum skeptic,” he says with a laugh, believing that an application to his work was “decades away.”
Challenges in Biotechnological Research
He and his team use big data and AI to discover proteins that could enable new immunotherapies more cheaply and quickly, often funded by the Novo Nordisk Foundation. While most biological modelers urgently need more data, for his team the lack of data on the full diversity of genetic information across the entire human race was particularly challenging because most medical research focused on Western populations. This can make it difficult to develop peptides that work in understudied populations like those in Asia and Africa, he says.
Quantum Computing’s Role in Diversity
His team hypothesized that embedding a quantum computer into its workflow could cause it to generate a more diverse set of peptides, particularly for targets for which there was less data, after learning that the machines had a similar effect in producing images.
Limitations and Future Prospects
The newly discovered method will not yet revolutionize research because quantum computers are still too small to run full-fledged, state-of-the-art AI models, meaning better results could be achieved on a classical computer.
“Quants are still not very powerful, so the complexity that we were able to encode was not that of a normal-sized antibody that we normally work with,” says DTU doctoral student Jonathan Funk. Furthermore, finding a peptide that can bind to a specific gene is only one step in vaccine development and would not alone lead to successful drugs.
Real-World Applications and Industry Skepticism
“I think it’s no surprise that many industrial companies think that quantum technology is unclear and a long way off,” Richard Murray, CEO of ORCA Computing, tells WIRED, in part because the technology “has never really had clear, near-term examples of its usefulness.”
He says this study is new in that it shows a near-term commercial application of quantum. His company is also applying the technology in projects with chemical oil giant BP and automaker Toyota to make its design process more efficient.
Future Research Directions
The DTU team will now examine whether they can use the workflow with state-of-the-art models and larger proteins. “We needed this to confirm in a simple way that we now actually have a chance to move the needle significantly,” says Patrick Jenkins, noting that generative AI workflows are particularly valuable in neglected diseases for which little research funding is available. He is also considering using a quantum computer to improve his generative AI method for developing synthetic antidotes to snakebite venom.
For more information, read the full article Here.
“`
