In brief
- Researchers identified a key molecular interaction that viruses rely on to enter cells and disrupted it in lab experiments.
- The work used AI and molecular simulations to narrow thousands of interactions down to one critical target.
- Scientists said the approach could help guide future antiviral and disease research, though it remains early-stage.
Most antiviral drugs target viruses after they have already slipped inside human cells. Researchers at Washington State University said they found a way to intervene earlier, identifying a single molecular interaction that viruses rely on to enter cells in the first place.
The research, published in November in the journal Nanoscale, focused on viral entry, one of the least understood and most difficult stages of infection to disrupt, using artificial intelligence and molecular simulations to identify a critical interaction within a fusion protein that, when altered in laboratory experiments, prevented the virus from entering new cells.
“Viruses attack cells through thousands of interactions,” Professor Jin Liu, a mechanical and materials engineering professor at Washington State University, told Decrypt. “Our research is to identify the most important one, and once we identify that interaction, we can figure out a way to prevent the virus from getting into the cell and stop the spread of disease.”
The study grew out of work that began more than two years ago, shortly after the COVID-19 pandemic, and was led by Veterinary Microbiology and Pathology Professor Anthony Nicola, with funding from the National Institutes of Health.
In the study, researchers examined herpes viruses as a test case.
These viruses rely on a surface fusion protein, glycoprotein B (gB), which is essential for driving membrane fusion during entry.
Scientists have long known that gB is central to infection, but its large size, complex architecture, and coordination with other viral entry proteins have made it difficult to pinpoint which of its many internal interactions are functionally critical.
Liu said the value of artificial intelligence in the project was not that it uncovered something unknowable to human researchers, but that it made the search far more efficient.
Instead of relying on trial and error, the team used simulations and machine learning to analyze thousands of possible molecular interactions simultaneously and rank which ones were most important.
“In biological experiments, you usually start with a hypothesis. You think this region may be important, but in that region there are hundreds of interactions,” Liu said. “You test one, maybe it’s not important, then another. That takes a lot of time and a lot of money. With simulations, the cost can be neglected, and our method is able to identify the real important interactions that can then be tested in experiments.”
AI is increasingly being used in medical research to identify disease patterns that are difficult to detect through traditional methods.
Recent studies have applied machine learning to predict Alzheimer’s years before symptoms appear, flag subtle signs of disease in MRI scans, and forecast long-term risk for hundreds of conditions using large health record datasets.
The U.S. government has also begun investing in the approach, including a $50 million National Institutes of Health initiative to apply AI to childhood cancer research.
Beyond virology, Liu said the same computational framework could be applied to diseases driven by altered protein interactions, including neurodegenerative disorders such as Alzheimer’s disease.
“The most important thing is knowing which interaction to target,” Liu said. “Once we can provide that target, people can look at ways to weaken it, strengthen it, or block it. That’s really the significance of this work.”
Generally Intelligent Newsletter
A weekly AI journey narrated by Gen, a generative AI model.
