A global coalition of researchers has just added 2,800 viral protein structures to the AlphaFold Database, an open-access resource that could help scientists prepare for the next pandemic. About 30% of these newly predicted structures have never been documented in scientific literature before, offering the global research community entirely new insights into how viruses function at the molecular level.

Why Are Scientists Stockpiling Viral Protein Data Now?

When COVID-19 emerged, scientists had a significant advantage: decades of prior research on coronaviruses meant they understood the virus's key proteins well enough to design vaccines in record time. The next pandemic may not offer the same head start. An analysis by the Center for Global Development estimates roughly a 50% chance of the world facing a pandemic as severe as COVID-19 by 2050.

The newly released dataset represents a deliberate effort to close that knowledge gap before the next outbreak occurs. The collaboration spans multiple organizations, including Google DeepMind, NVIDIA, the European Molecular Biology Laboratory's European Bioinformatics Institute (EMBL-EBI), Seoul National University, Sungkyunkwan University, the Swiss Institute of Bioinformatics, and the University of Glasgow.

"When the next pandemic happens, there may be something that comes out of the blue, and we'll be lacking the knowledge we had for COVID. What we're trying to do is stockpile some of that knowledge ahead of time," said Joe Grove, professor of molecular virology at the Medical Research Council-University of Glasgow Centre for Virus Research.

Joe Grove, Professor of Molecular Virology, Medical Research Council-University of Glasgow Centre for Virus Research

How Does AlphaFold Predict Protein Structures So Quickly?

Traditional methods for determining protein structures involve crystallizing proteins and shooting X-rays at them, a process that can take years and cost thousands of dollars per structure. AlphaFold2, Google DeepMind's artificial intelligence model for predicting how proteins fold into three-dimensional shapes, predicts a structure in minutes and can be run in bulk.

For this project, the team systematically worked through the protein structures of viral families known to infect humans, from common-cold viruses to emerging threats like Mpox. The structures were inferred using AlphaFold2 with optimization from NVIDIA BioNeMo Inference Runtime, a GPU-accelerated tool that allowed the team to scale inference to thousands of viral proteomes.

Most proteins don't work alone. They come together in complexes of multiple molecules to perform sophisticated functions, and those structures are often what a vaccine or drug must target to disrupt viral function. Understanding the three-dimensional structure of the COVID-19 virus's spike protein, for example, proved foundational to vaccine design.

Steps to Access and Use the New Viral Protein Dataset

• Explore the Database: Scientists can access the viral protein complex dataset directly through the AlphaFold Database Pandemic Preparedness Portal, where all predictions are labeled by confidence level to help researchers assess reliability.
• Run Your Own Predictions: NVIDIA is openly releasing the BioNeMo Structure Prediction Pipeline, the GPU-accelerated workflow used to generate the dataset, allowing researchers to predict three-dimensional structures for their own protein targets starting from just a protein sequence.
• Verify High-Confidence Predictions: Scientists can use the open dataset as a starting point and then verify high-confidence predictions through experimental methods in their own laboratories, combining computational speed with physical validation.

The AlphaFold Database now holds more than 260 million protein and protein complex predictions covering nearly every cataloged protein known to science. This expansion represents a major contribution to the information available for scientists across digital biology and disease research.

"Our ambition with the AlphaFold Database has always been to democratize access to foundational biology at scale. This collaboration to bring thousands of viral complexes into the database will equip scientists around the world with insights they need to help prepare for future outbreaks," said Risha Patel, life sciences partnerships manager at Google DeepMind.

Risha Patel, Life Sciences Partnerships Manager, Google DeepMind

What Makes This Dataset Unique for Low-Resource Settings?

One of the most significant aspects of this release is its accessibility. By making the data openly available, the collaboration lowers barriers for scientists in low-resource settings who are confronting outbreaks firsthand. Traditional structural biology methods require expensive equipment and years of training, putting them out of reach for many researchers in developing nations.

The dataset also covers lesser-studied viruses, not just the high-profile pathogens that attract research funding. This breadth means that scientists working on emerging threats in regions with limited resources can access the same foundational knowledge as researchers at well-funded institutions in wealthy countries.

"When I did my Ph.D., there were no structures for any of the proteins we were investigating. It was like working in the dark. We had to guess what was going on. This dataset is a powerful tool for all the researchers doing their Ph.D.s now, giving them high-quality structural data that's going to accelerate fundamental science," noted Joe Grove.

Joe Grove, Professor of Molecular Virology, Medical Research Council-University of Glasgow Centre for Virus Research

The dataset release coincided with a United Nations General Assembly meeting convened by the World Economic Forum on pandemic prevention, preparedness, and response taking place in New York City. This timing underscores the global health significance of the effort and reflects a broader shift toward using artificial intelligence not just for commercial applications but for pandemic preparedness and public health.

As the global research community continues to expand AlphaFold's capabilities, the focus on viral proteins represents a strategic investment in collective preparedness. By making this knowledge openly available now, scientists are laying groundwork that could accelerate vaccine and treatment development when the next outbreak occurs, potentially saving millions of lives.