University news

Harnessing the power of AI, a research team at the MRC-University of Glasgow Centre for Virus Research has launched Viro3D – the most comprehensive database of human and animal virus protein structure predictions in the world.

A free and searchable AI-powered database, Viro3D offers a completely new, in-depth perspective on viruses, allowing us to learn more and faster about their origins and evolution.

Although virus particles are the most abundant biological entities on our planet, these tiny structures remain among the least well-understood. Insights into the key protein structures within viruses have, until now, only been achieved through slow and laborious research work, a pace that has impacted our ability to develop treatments and vaccines at speed.

Now, in an effort to accelerate our understanding, a team of researchers, led by Dr Joe Grove at the MRC-University of Glasgow Centre for Virus Research, has harnessed the power of AI to create a new database of thousands of human and animal virus proteins.

Viro3D contains high-quality structural models for 85,000 proteins from 4,400 human and animal viruses – the largest database of complete structural models for human and animal viruses, expanding our current knowledge in this area by 30 times.

The creators of Viro3D expect the database will begin a new era for human and animal virus research, and accelerate the development of antiviral drugs and vaccines against existing endemic viruses, such as hepatitis, HIV, the common cold and COVID-19, as well as for new and emerging pandemic threats.

The database, which was recently detailed in the journal Molecular Systems Biology, has already revealed some previously unknown information on the genetic ancestry of SARS-CoV-2, the virus responsible for COVID-19. Data from Viro3D suggests that a key protein in SARS-CoV-2, one of the ones responsible for infection, may have originally come from a genetic exchange with an ancestral herpesvirus, the family of viruses responsible for a range of illnesses including cold sores and chickenpox.

The research team say Viro3D has the potential to generate much more information like this – on the evolutionary history and origins of viruses – for thousands of different proteins, helping to turbocharge our understanding of human and animal viruses, and in turn our ability to combat them at pace.

Dr Joe Grove, lead author from the MRC-University of Glasgow Centre for Virus Research, said: “We are only at the beginning of understanding the enormous genetic diversity within the viral community. Viruses are as old as cellular life, having been around for billions of years.

“Viro3D provides a completely new perspective on viruses, allowing us to learn about their origins and evolution. We expect Viro3D, and artificial intelligence, to accelerate the computational design of antiviral drugs and vaccines against existing endemic viruses. We can then bring these designs back into laboratories and then to the real world.”

Proteins are the building blocks of life, performing a huge range of functions for our cells and playing critical roles in both health and disease. In order to understand how proteins work, researchers must be able to examine their three-dimensional structure.

While advances in artificial intelligence have led to the creation of general protein structure databases, such as The AlphaFold Database and The Evolutionary Scale Modelling Metagenomic Atlas, virus proteins have so far either been excluded from prediction efforts or only included in small numbers.

Viro3D is detailed in the study ‘Viro3D: a comprehensive database of virus protein structure predictions’. The work was funded by Wellcome, The Medical Research Council (MRC) and the Medical Research Foundation.

---

Enquiries: ali.howard@glasgow.ac.uk or elizabeth.mcmeekin@glasgow.ac.uk

First published: 26 September 2025