Researchers at Washington State University have made significant strides in the fight against viral infections by employing artificial intelligence (AI) to identify a crucial molecular interaction that facilitates viral entry into human cells. This groundbreaking work was published in the journal Nanoscale in November and represents a promising shift in antiviral research.
Traditionally, most antiviral drugs target viruses after they have already invaded human cells. However, the team at Washington State University has developed a method to intervene at an earlier stage, focusing on the entry mechanism of viruses. By isolating a specific interaction within a fusion protein, the researchers were able to disrupt the process that allows viruses to enter new cells, thereby halting the spread of disease.
Professor Jin Liu, who specializes in mechanical and materials engineering, stated, “Viruses attack cells through thousands of interactions. Our research is to identify the most important one. Once we pinpoint that interaction, we can devise strategies to prevent the virus from invading the cell.”
The study is the culmination of over two years of research initiated in response to the COVID-19 pandemic. Lead by Professor Anthony Nicola from Veterinary Microbiology and Pathology, this project received funding from the National Institutes of Health.
In their experiments, the researchers focused on herpes viruses, which utilize a surface fusion protein known as glycoprotein B (gB) to facilitate membrane fusion during cell entry. Although gB has long been recognized as vital for infection, its complex structure and interactions with other viral entry proteins have made it challenging to determine which specific interactions are critical.
The application of AI significantly enhanced the efficiency of their research. Rather than relying on traditional trial-and-error methods, the team harnessed simulations and machine learning algorithms to evaluate thousands of potential molecular interactions simultaneously, identifying the most critical ones for further investigation.
“In biological experiments, you usually start with a hypothesis,” Liu explained. “You think a particular region may be important, but within that region, there are hundreds of interactions to test. This can be time-consuming and expensive. Our simulations allowed us to identify key interactions quickly and effectively, making it feasible to test them in the lab.”
The use of AI in medical research is gaining traction as it helps detect complex disease patterns that traditional methods may overlook. Recent applications of machine learning have successfully predicted Alzheimer”s disease years before symptoms manifest and identified subtle anomalies in MRI scans. The U.S. government has also begun funding AI initiatives, including a $50 million commitment from the National Institutes of Health aimed at applying AI technologies to childhood cancer research.
Professor Liu emphasized that the computational framework developed during this research could extend beyond virology, potentially aiding in the understanding and treatment of various diseases linked to altered protein interactions, such as neurodegenerative conditions like Alzheimer”s disease. “The critical aspect is identifying which interaction to target,” Liu remarked. “Once we establish that target, researchers can explore ways to weaken, strengthen, or block it, thereby advancing our understanding and treatment of various diseases.”
