Artificial intelligence is helping scientists anticipate how viruses will evolve, enabling them to design vaccines that stay ahead of the virus's shape-shifting defenses. Researchers at Iowa State University are using AI-powered simulations to predict viral mutations and develop multi-epitope vaccines (vaccines targeting multiple parts of a virus) that protect livestock against numerous adaptations of diseases like porcine reproductive and respiratory syndrome (PRRS), which costs the global pork industry more than $1 billion annually .

Why Are Livestock Viruses So Hard to Vaccinate Against?

The PRRS virus is one of the fastest-evolving RNA viruses known to infect animals. As soon as an animal's immune system learns to recognize and neutralize one variant, the virus mutates into a new form that evades detection. This evolutionary arms race makes traditional vaccine development feel like chasing a moving target. The virus essentially "shape shifts," trying to create new molecular structures that won't trigger an immune response .

Understanding how viruses attach to cells and infect them is crucial to vaccine design. Researchers identify specific protein segments that antibodies can recognize and neutralize, called epitopes. But if the virus changes the shape of these protein segments even slightly, the vaccine becomes ineffective. As one researcher explained, the challenge is comparable to building with Legos, except these are "soft Legos" where tiny changes can dramatically alter the structure .

How Does AI Speed Up Vaccine Design?

Ratul Chowdhury, an assistant professor of chemical and biological engineering at Iowa State, leads a research group using AI to screen millions of possible protein shapes and mutations to identify which ones could help viruses avoid immune detection. This computational approach is approximately 100 times faster than traditional methods, making it feasible to analyze the vast complexity of viral evolution .

Chowdhury's team developed a digital platform called AstraMEV for infectious bronchitis virus (IBV), a common chicken pathogen. The platform analyzed 56 clusters of similar viral variations from public sequencing data to identify which genetic snippets remain stable as the virus evolves and which help it infect cells. From 258 epitopes identified in that analysis, researchers designed two types of multi-epitope vaccines: SmallTope (containing up to eight epitopes) and BigTope (containing up to 23 epitopes) .

For PRRS specifically, Chowdhury's team identified 75 epitopes and designed 56 immunogens, short protein fragments that could serve as potential vaccines. These findings were published in the Computational and Structural Biotechnology Journal in late 2024 .

Steps to Translating AI Vaccine Research Into Commercial Products

• Industry Partnership: Iowa State's chief technology officer for vaccines and immunotherapies, Mike Roof, a former global segment head of vaccine development at Boehringer Ingelheim, works directly with researchers to ensure studies address real industry problems and meet regulatory requirements.
• Live Animal Testing: Chowdhury's lab is planning a small trial of a PRRS vaccine at an Iowa State swine farm later this year to validate AI-designed vaccines in real-world conditions.
• Commercialization Infrastructure: CYVAX, a small-scale wet lab in the ISU Research Park, provides flexible and cost-effective space for federally compliant clinical trials, potentially reducing development time from several years to just a few months.
• Data Advantage: Iowa State's Veterinary Diagnostic Laboratory (VDL) conducted more than 1.7 million tests last year, providing the rich animal health datasets that make AI models more effective and accurate.

Multiple animal health companies, ranging from large corporations to small startups, have already told Chowdhury they plan to test his published vaccine research in their own laboratories .

Why Does Iowa State Have an Edge in This Research?

Before joining Iowa State in 2022, Chowdhury was a postdoctoral researcher at Harvard Medical School studying computational structural biology for cancer and pain treatment. He observed that in human health research, pharmaceutical companies typically control the most valuable datasets. In animal health, however, universities often lead. Iowa State's advantage is substantial: the VDL has the largest U.S. livestock caseload, providing researchers with unparalleled access to detailed animal disease data .

"If your data set is detailed, your methods don't need to be funky. We have the benefit of having rich data for understanding patterns, and that puts us in the driver's seat for performing cutting-edge research," said Ratul Chowdhury, assistant professor of chemical and biological engineering at Iowa State University.

Ratul Chowdhury, Assistant Professor of Chemical and Biological Engineering at Iowa State University

The ongoing stream of anonymized data from the VDL provides timely inputs for AI models, enabling researchers to identify emerging patterns in viral evolution and refine vaccine designs accordingly .

What Support Systems Enable This Research?

Iowa State's infrastructure extends beyond data. The state of Iowa supports four biosciences platforms designed to transform existing research strengths into innovation-powered economic growth. Chowdhury's work benefits from proximity to industry expertise, with Mike Roof's office located just one door down from his at the Nanovaccine Institute. This arrangement allows researchers to receive mentorship on translating academic findings into commercially viable products .

"Sometimes a little tweak or a slight change in study design can make the results more appealing to industry or for commercial success. For faculty with strong academic backgrounds but maybe not much industry experience, I can mentor them and help accelerate their work," said Mike Roof, Iowa State's chief technology officer for vaccines and immunotherapies.

Mike Roof, Chief Technology Officer for Vaccines and Immunotherapies at Iowa State University

Chowdhury emphasized that his research success depends on this broader ecosystem. "This isn't a one-man show," he noted. "There's so much supportive infrastructure here" .

Chowdhury

As animal health companies begin testing these AI-designed vaccines in their own laboratories, the approach could fundamentally change how the livestock industry responds to viral threats. Rather than waiting for viruses to mutate and then scrambling to develop new vaccines, researchers may soon be able to anticipate evolutionary changes and deploy protective measures before new variants cause widespread disease.