The biotechnology and pharmaceutical firm Moderna Inc., best known for its COVID-19 vaccine, said today it’s working with IBM Corp. to explore ways in which quantum computing and generative artificial intelligence can advance its mRNA technology.

The two companies have signed an agreement that will allow Moderna’s scientists to access IBM’s quantum computing systems, which most experts agree are among the world’s most advanced. The quantum systems can help Moderna to accelerate the discovery and creation of new messenger RNA vaccines and therapies, said Dario Gil, director of IBM Research.

Cambridge, Massachusetts-based Moderna became a household name after successfully developing one of the first mRNA vaccines during the coronavirus pandemic. mRNA works by teaching human cells to produce a protein that initiates an immune response against specific diseases. As demand for COVID-19 vaccines wanes, Moderna is trying to harness mRNA technology to create medicines for other kinds of diseases.

That’s where IBM believes it can help. As part of the arrangement announced today, it will provide quantum computing experts that can help Moderna’s scientists implement the technology for their very specific needs, the companies said. Quantum machines are very different from classical computers, which store information in either zeroes or ones. Instead, quantum computers are based on quantum physics, and that allows them to solve problems that are far too complex for traditional machines.

The great promise of quantum computing is that it can model the attributes and behaviors of molecules more accurately. That’s because the molecules themselves fundamentally operate according to the principles of quantum mechanics. Quantum computers, which use quantum algorithms that leverage the entanglement between qubits, are therefore far better suited to capture and replicate the behavior of molecular systems.

In addition to the quantum systems, Moderna’s researchers will also access IBM’s foundational generative AI model, called MoLFormer. Generative AI refers to algorithms that are able to generate new content based on the data they were trained on.

According to IBM, it’s possible to train generative AI models on broadscale molecular datasets, and then adapt them to different use cases and tasks. In the case of MoLFormer, it’s said to have been trained on a broad knowledge base of more than one billion chemicals, meaning it already has a solid understanding of chemistry and how different molecules will interact. Moderna said it will use MoLFormer to try and understand the characteristics of potential MRNA medicines.

Foundation models like MolFormer are key to accelerating chemical insights, said Holger Mueller, an analyst with Constellation Research Inc. “They move the experimentation from the lab to the digital realm,” he explained. “It’s interesting that Moderna is looking to combine this with quantum computing technology, because quantum is the ideal platform to model chemical and molecular processes. It can do this much more efficiently than existing platforms, so we can expect to see scientific progress result from this partnership.

Charles King of Pund-IT Inc. agreed that complex chemical processes, including drug discovery and genomics, are among the richest areas targeted by quantum-enabled platforms. “Moderna’s focus on mRNA-based vaccines and research, plus IBM’s leading edge quantum and AI innovations, make the pair logical allies in next-generation product exploration and development,” he said.

The intersection between generative AI and quantum computing is not the main focus at this time, but it is an area that Moderna and IBM intend to explore to see if and how the two technologies can be used together to aid in drug discovery.

IBM has been expanding its investment in AI lately, with a particular focus on partnerships such as the one announced today. Earlier this year, it agreed a deal with the National Aeronautics and Space Administration to help build foundational AI models that can help to advance climate science.

IBM’s efforts come at a time when many technology firms are engaged in an AI arms race, driven by the release of OpenAI LP’s ChatGPT, which is built on generative AI models. ChatGPT is able to answer questions in a clear, concise way, similar to how an educated human might respond, and quickly caused a sensation after it was launched.

Photo: Rafael Classen