The Industrial Generative AI company Zapata Computing, Inc. (hereinafter referred to as “Zapata AI” or the “Company”) today announced that its scientists, working with Insilico Medicine, the University of Toronto, and St. Jude Children’s Research Hospital, have produced the first example of a generative model operating on quantum hardware that produces viable cancer drug candidates more quickly than state-of-the-art classical models. According to the research, hybrid quantum generative AI for drug development employing current quantum devices has a bright future.
In the study, generative AI was used by the researchers to create new KRAS inhibitors, a crucial area of research in cancer treatment that has previously been thought to be “undruggable” because of its inherent biochemical characteristics. One million drug candidates were produced by generative models operating on classical hardware, quantum hardware (a 16-qubit IBM device), and simulated quantum hardware. The drug candidates were subsequently screened both manually and algorithmically. After that, 15 different compounds were created and examined using tests based on cells. The two compounds produced by the quantum-enhanced generative model had a higher binding affinity than the molecules produced by strictly classical models and were unique from KRAS inhibitors that were already on the market.
Yudong Cao, CTO and co-founder of Zapata AI, said, “This project is an exciting demonstration of how quantum and classical computing can complement each other to deliver an end-to-end solution.” Another excellent illustration of how the startup and academic ecosystems may benefit from one another’s capabilities to advance advancement is the partnership between Zapata, UofT, St. Jude, and Insilico. We are eager to advance this study in order to advance the identified compounds through the drug development process, apply our approach to more disease targets, and expand the applications of our quantum-enhanced generative AI to additional industrial use cases including challenging design problems.
While it waits for peer review, the study is presently available for download on ArXiv. The research is a follow-up to a 2023 study that the team and Foxconn published that first demonstrated the potential of quantum generative AI for drug discovery.
According to Alex Zhavoronkov, PhD, the founder and co-CEO of Insilico Medicine, “this research provides further validation of the potential of Insilico’s generative AI engine, Chemistry42, to be combined with quantum-augmented generative models in order to develop novel therapeutic possibilities for difficult-to-drug targets in cancer and other indications.” “With Zapata AI and Alán Aspuru-Guzik at the University of Toronto, we look forward to working together to further develop these methods, as this represents an important first step toward a more advanced drug discovery future.”
The revelation also comes after it was recently announced that D-Wave Quantum Inc. (NYSE: QBTS) (“D-Wave”) and Zapata AI have formed a new strategic alliance. The first goal of this alliance is to develop quantum generative AI models that will speed up the identification of novel compounds for use in commercial applications. Christopher Savoie, CEO and co-founder of Zapata AI, stated, “With quantum-enhanced generative AI, we’ve been able to produce real effective drug lead molecules for the first time ever.” The nice thing is that this is just the start. This is the same technology that we are working with D-Wave to develop into a commercial product, which we anticipate bringing to market soon due to the advanced commercial maturity of both D-Wave’s annealing quantum computing and Zapata’s generative AI technology. We’re excited to carry out more study on this to find novel compounds for pharmaceutical development and other industrial uses.
Alán Aspuru-Guzik, a co-founder and scientific advisor of Zapata AI in addition to being a professor of chemistry and computer science at the University of Toronto, said, “I have always been excited about the potential of AI and quantum computing for drug and materials discovery.” The integration of quantum computing modules into the drug discovery pipeline is still in its early stages of development. The fact that we were able to identify a novel chemical that inhibits KRAS is fantastic. There are a lot of open-ended inquiries. It is thrilling to see that this study paves the way for future, more powerful quantum computers to demonstrate their capabilities, even if everything in this paper could also be done with a classical computer. Researchers from all across the world will be able to build on this groundbreaking experiment.
The study made use of the QML Suite Python Package on the Orquestra® platform of Zapata AI, which is accessible online.
