WIREDGOV

Researchers at STFC's Hartree Centre teamed up with IBM Research and Salient Bio, to unlock new insights into one of the world's most common chronic conditions.

Thanks to artificial intelligence (AI), their teams findings could change how gum disease is diagnosed and treated in the future.

Working with microbiome testing and diagnostics company Salient Bio and IBM Research, the team have developed a new AI-powered approach to understanding periodontitis.

This is a severe form of gum disease that affects hundreds of millions of people globally.

A more complete picture

For years, diagnosing gum disease has relied on checking for the presence or absence of specific harmful bacteria.

However, this approach misses something crucial.

It’s not one rogue bacterium that causes the problem, but a breakdown in the delicate balance of the entire community of microorganisms living in the mouth.

Exceptionally complex

The human mouth is exceptionally complex, harbouring the second most diverse microbial ecosystem in the body after the gut.

It contains over 700 species of bacteria, alongside fungi and viruses.

Together, the team analysed genetic data from the mouth bacteria of 600 samples, mapping out how different species interact with one another.

Using machine learning, they identified tightly connected clusters of bacteria that act together as units, and crucially, which clusters are linked to healthy mouths and which to diseased ones.

13 species, one powerful diagnostic

From hundreds of variables, AI distilled the complexity of the entire microbiome down to just 13 key bacterial species.

This compact panel proved just as accurate at diagnosing gum disease as models relying on vastly more data, a significant step towards a practical, targeted diagnostic tool.

The research also factored in lifestyle and environmental data from patients, helping to build a richer, data driven picture of why some people are more at risk than others.

Stopping disease before it starts

The real promise of this work lies in prevention.

By identifying individuals at risk before their symptoms become severe, clinicians can intervene earlier, using targeted therapies rather than the broad-spectrum antibiotics that are typically prescribed today.

This is particularly important at a time when antibiotic resistance is a growing global concern and the healthcare system is shifting towards preventative care measures.

Bringing together expertise

Tom Sewell, Lead Bioinformatician at Salient Bio, said:

Collaborating with the Hartree Centre and IBM Research helped us to bring together a great set of expertise to help us find association between bacteria in the microbiome and disease which has huge applications across the healthcare system.

We went from oral sample collection through metagenomic sequencing to a validated 13-species diagnostic signature. That wouldn’t have been possible without that partnership.

Beyond gum disease

The implications have the potential to stretch further than dentistry.

This framework could serve as a blueprint for understanding other chronic healthcare conditions linked to the human microbiome, opening doors to new diagnostics and therapies across a range of diseases.

Powering UK industry with AI

Located at STFC’s Daresbury Laboratory, at Sci-Tech Daresbury in the Liverpool City Region, the Hartree Centre is the UK’s only supercomputing centre dedicated to industry engagement.

The research with Salient Bio was carried out through the Hartree National Centre for Digital Innovation (HNCDI), a five-year, government-funded collaboration between STFC and IBM Research.

It was designed to equip businesses with skills and technical knowledge to adopt emerging digital technologies.

AI for disease prevention

Kate Royse, Hartree Centre Director, said:

At the Hartree Centre, our mission is to equip UK industry with the knowledge, skills and compute needed to fully unlock the potential of advanced digital technologies.

This research with Salient Bio is an excellent example of how AI can play a powerful role in preventative medicine, helping us understand and address disease before it takes hold, rather than simply treating it.

It is an excellent example of how some of the most powerful breakthroughs in medicine don’t always come from laboratories alone. Sometimes, they come from combining biological expertise with the kind of advanced computing and AI capabilities that the Hartree Centre provides.