AI unlocks RNA’s role in Dementia and emerging viral threats

To solve a problem, scientists first need to see it clearly. Whether it’s a virus slipping past the immune system or plaques forming in the brain, visualization is the first step toward finding answers. Yet this process is often both difficult and expensive.

A breakthrough in artificial intelligence is helping to change that. Researchers have developed a powerful new tool that creates more accurate 3D models of molecular interactions inside the body. The method is poised to speed drug discovery and sharpen our understanding of complex diseases.

The tool, called ProRNA3D-single, was created by computer scientist Debswapna Bhattacharya and his team. It represents a leap forward in predicting how viral RNA interacts with human proteins, a process central to many illnesses.

“The ultimate goal is to accelerate the drug discovery process to prevent the RNA viruses from interacting with host proteins, potentially stopping infections before they grow into pandemics or inhibiting altered function of RNA binding proteins in Alzheimer’s disease,” Bhattacharya explained.

For decades, researchers have struggled to capture how ribonucleic acid molecules twist, fold, and bind to proteins in three dimensions. These shapes decide whether a virus like SARS-CoV-2 can replicate or whether Alzheimer’s disease gains a foothold. ProRNA3D-single makes those hidden interactions visible with remarkable accuracy.

To create the system, the Virginia Tech team drew inspiration from language models such as ChatGPT. Instead of analyzing human words, however, their models learn the “alphabet” of DNA, RNA, and proteins. These models can then be trained to predict how molecules behave.

“The bio LLMs are basically like ChatGPT, but for biological sequences. And just like ChatGPT, we can ask our models questions and get answers,” Bhattacharya said.

The team built a model where separate large language models for proteins and RNA “talk” to each other. From those exchanges, ProRNA3D-single generates 3D structures showing viral RNA docking onto proteins in human cells. “This is basically a neural pairing of two different large language models, leading to bilingual reasoning,” Bhattacharya added. “From a computer science standpoint, that’s a contribution in itself.”

Even the most advanced AI programs from groups such as Google DeepMind have stumbled when trying to model RNA-protein complexes. By contrast, the Virginia Tech approach provides far more accurate structures, saving scientists from expensive trial-and-error experiments.

The accuracy of ProRNA3D-single is more than a technical achievement. It has practical implications for medicine and public health. Diseases from COVID-19 to dementia may soon be better understood at the molecular level.

Instead of relying on guesswork, researchers can now study how viruses attach to proteins and design drugs to block those sites. This ability could slash both the cost and the time needed to respond to new outbreaks.

“If you remember the COVID-19 pandemic and the mRNA-based vaccine that actually helped a lot — that vaccine worked because it was an RNA-based therapeutic,” said Ph.D. student Sumit Tarafder. “Modeling of protein-RNA interactions in 3D is crucial, so that we know where the drug can actually target molecules that cause disease.”

The promise extends well beyond viruses. By mapping interactions between RNA and proteins, the model may also uncover pathways tied to neurological decline and other illnesses that remain difficult to treat.

The development of ProRNA3D-single took two years and the efforts of several researchers. Doctoral student Rahmatullah Roche, now a faculty member at Columbus State University, contributed key coding work and published more than a dozen related studies during his graduate career. “The lead Ph.D. students did enormous work,” Bhattacharya said. “They did most of the heavy lifting.”

The project was funded by the National Institutes of Health and the National Science Foundation. In keeping with the spirit of open science, both the paper and the tool itself are freely available for researchers worldwide.

“We can’t overstate the importance of investing in science to benefit society. We believe that openness is the key to making science accessible to everybody,” Bhattacharya said. “Taxpayers fund us, so we have an obligation to give back, which is why we make our work open source and publicly available.”

The team plans to refine ProRNA3D-single further, improving its accuracy and expanding its applications. “We should constantly remind ourselves the problem is far from being solved,” Bhattacharya said. “We made progress, yes, but we’re mindful of the fact that these models have a long way to go.”

Tools like ProRNA3D-single represent more than just computational triumphs. They reflect a shift in how humanity approaches biology. Instead of spending years on costly lab experiments, scientists can now pair AI with bench science to accelerate discovery.

The vision is clear: faster vaccines, better therapies, and a deeper grasp of how diseases work. As the world faces future pandemics and the rising burden of neurological disorders, breakthroughs like this may prove indispensable.

Research findings are available online in the journal Cell Systems.

Like these kind of feel good stories? Get The Brighter Side of News' newsletter.