New $18M Penn project will use AI to develop RNA treatments like the COVID vaccines

Cellular-level biology that could help develop vaccines just got more accessible to researchers, thanks to AI. 

The University of Pennsylvania, along with the University of Puerto Rico, received an $18 million six-year grant from the National Science Foundation (NSF) earlier this fall to support an artificial intelligence-based RNA “BioFoundry.” 

“The goal is to democratize RNA technology,” Daeyeon Lee, engineering biomolecular professor at Penn, told Technical.ly. “So that if there’s a scientist that’s sitting in their lab and they’re trying to solve a problem, [RNA is] one of the tools in [their] toolbox.” 

RNA, or ribonucleic acid, is a molecule in cells that carries genetic information. For example, it was used in recent years to develop the COVID vaccine and make other advancements in healthcare, but it can also be used to address challenges in agriculture and environmental sustainability, according to Lee.

The acid functions similarly in all living organisms, Lee said. If it’s easier to access, more people can use it to find treatments for diseases or protect access to food. The BioFoundry is currently developing an AI-based platform for researchers to connect on called AIRFoundry Knowledge Sharing Platform, Lee said. 

The BioFoundry aims to help scientists use AI to synthesize the research that’s out there, share it with other researchers and track their progress. The goal is to create a computer interface with AI tools that researchers can use to answer their questions or that will put them in contact with subject experts, according to Lee. 

The BioFoundry will also provide physical materials, such as nanoparticles, the material used to transport RNA into cells. 

Expanding RNA research for medical innovations and beyond 

The BioFoundry is especially interested in RNA because the current research community around it is very small, said Zachary Ives, computer and information science department chair at Penn.

RNA can change cellular behavior, generate proteins and be used for genetic editing, Ives said. 

The BioFoundry will start by working with experienced RNA researchers and learn what the process currently looks like. Then, the scientists will build a foundational AI system that reflects that process, Ives said. They will eventually expand to researchers from a variety of institutions and fields, and continue to update the AI system with their work. 

Right now, RNA research is mostly focused on human health applications. The goal is to make it more reasonable for RNA to be used in the context of non-human living organisms. 

For example, it can be used to develop enzymes that break down pollutants like plastic or to protect crops against pests or harsh weather. 

“We have to build something that goes end to end,” he said. “That is a combination of tools from different fields and common practices.”

AI summarizes vast knowledge to lower the barrier to entry

As a part of the BioFoundry, artificial intelligence will be used to guide researchers who are just getting into the field, providing expert knowledge they wouldn’t have access to otherwise, Ives said. 

“Broadly, we want to have mechanisms that help somebody move into the field and make it not a specialized thing,” Ives said. “But something that becomes prevalent across the life sciences.”

The NSF grant allows the BioFoundry to keep user fees low. There will also be grant opportunities for researchers looking to do specific projects with its resources, Lee said. 

The BioFoundry will also use AI to track impact, by keeping records of who comes to it, what information they request and what materials they request. 

The professors also want to keep track of research results, like whether they led to published reports or commercial products, Ives said. 

“Then the question is, do they come back to the BioFoundry so that they can tell us what their results were,” Lee said. “Once they do that, they’re also contributing to the BioFoundry, to the AI component.”