Breaking New Ground: How AI and Protein Patterns Are Changing Disease Prediction

Leveraging AI and machine learning to predict Alzheimer’s, diabetes, and heart disease up to 10 years in advance

Add bookmark
Joanna Edwards
Joanna Edwards
08/14/2024

Visual representation of AI and protein pattern analysis in predicting diseases like Alzheimer’s and diabetes

Introduction:

Imagine being able to predict your risk of developing Alzheimer’s, type 2 diabetes, or heart disease a decade before any symptoms appear. Thanks to a groundbreaking study led by researchers from Optima Partners, Biogen, and the University of Edinburgh, this possibility is closer to reality than ever before. Optima Partners, a leading data and business consultancy known for its work in advanced data science, joined forces with Biogen, a global biotechnology leader focused on neurological diseases, and the University of Edinburgh, renowned for its cutting-edge research in molecular epidemiology. Dr Chris Foley, Managing Director and Chief Scientist at Optima Partners, played a crucial role in guiding the research, while Dr Danni Gadd, the first author of the study and a neuroscientist with a strong background in computational research, provided key insights into the findings. Dr Riccardo Marioni, Chair of Molecular Epidemiology of Ageing at the University of Edinburgh and also an author of the paper, contributed his expertise in linking bio-samples to future health outcomes. By analysing vast amounts of medical data from the UK Biobank using advanced AI techniques, these researchers have discovered specific protein patterns—known as protein signatures—that can accurately predict the risk of these diseases, sometimes as much as 10 years before a diagnosis.

We had the opportunity to speak with Dr Chris Foley and Dr Riccardo Marioni in an interview where they shared their insights into this groundbreaking research. 

Background: Understanding the Power of Machine Learning in Disease Prediction

This study is particularly exciting because it shows how machine learning can delve deep into the data to reveal connections that were previously invisible. The research involved nearly 50,000 participants who provided blood samples between 2006 and 2010. By examining these samples, the team identified protein signatures that correlate with the risk of developing certain diseases.

Dr Riccardo Marioni, Chair of Molecular Epidemiology of Ageing at the University of Edinburgh and one of the study’s authors, explained the essence of the research: “We’re looking at what a sample of blood can tell us about your future health. The exciting part is finding links between proteins in the blood and the risk of brain-based diseases like Alzheimer’s. It’s a step forward because, traditionally, the thought was that you needed to look at the brain itself to understand these diseases.”

Key Findings: Protein Signatures Offer Early Warning for Diseases

One of the most remarkable findings from the study was that existing blood tests, when analysed alongside these newly identified protein patterns, could provide early warnings of disease risk. This could be a game-changer for clinicians who, armed with this information, could take preventive measures much earlier.

GDF15, a protein associated with inflammation, was found to be particularly telling. It showed links to nearly half of the 23 diseases studied, including Alzheimer’s, vascular dementia, heart disease, liver disease, type 2 diabetes, and even all-cause mortality. Dr Danni Gadd, the study’s first author, noted in the original press release, “Our research represents a promising step forward in risk prediction. Being able to detect early warning signs for a broad set of conditions may lead to opportunities for early intervention and prevention, marking a significant moment for the healthcare industry.”

Clinical Implications: A New Era of Personalised Medicine

The practical implications of these findings are profound. Imagine a world where doctors can tell you not only that you’re at risk for a particular disease but can also offer tailored advice to help you avoid it. Dr Chris Foley, Managing Director and Chief Scientist at Optima Partners, discussed this potential: “What we’ve done here is lay the groundwork. We’re not just predicting risk; we’re potentially enabling a whole new level of personalised medicine. But, of course, more work is needed to bring these findings into everyday clinical practice.”

Dr Foley also highlighted the importance of the collaboration between academia and industry in making these advancements possible. “It’s about bringing the best of both worlds together,” he said. “We need the cutting-edge research that universities like Edinburgh are known for, combined with the agility and commercial insight that companies like Optima and Biogen provide.”

Expert Insights: Collaboration Driving Breakthroughs in Disease Prediction

During our conversation, Dr Marioni and Dr Foley both emphasised how collaboration was key to the success of this study. Dr Foley remarked, “In health sciences, especially in early-phase drug discovery, bridging the gap between academia and industry is crucial. What we’ve done with this study is a perfect example of how these partnerships can lead to real breakthroughs.”

Dr Marioni echoed this sentiment, adding, “For us, it’s been fantastic to work with Optima and Biogen. It’s not just about sharing data—it’s about sharing expertise, too. We’re bringing different strengths to the table, and that’s what makes the whole greater than the sum of its parts.”

Future Directions: Where Does Disease Prediction Go From Here?

The study, which was published in Nature Aging (July 2024), is just the beginning. The researchers are already thinking about the next steps—how to refine these findings, validate them across different populations, and, ultimately, bring them into the clinic. You can access the full study here: Blood protein assessment of leading incident diseases and mortality in the UK Biobank.

Dr Foley mentioned, “The big challenge now is taking what we’ve learned and applying it in a way that’s meaningful for patients. We’re talking about moving from research to real-world applications, and that’s where the hard work really begins.”

Dr Marioni added, “There’s also the question of making sure these findings are applicable to everyone, not just specific groups. We need to ensure that the AI models we’re developing are robust and fair across different populations. It’s about equity in healthcare, which is more important now than ever.”

Conclusion: A Leap Forward in Disease Prediction and Prevention

This research is a major leap forward in the field of disease prediction. By harnessing the power of AI and protein pattern analysis, we’re moving closer to a future where diseases can be prevented before they even begin. It’s an exciting time for healthcare, with the potential for significant improvements in how we manage and prevent chronic diseases.

For those in the pharmaceutical and biopharmaceutical industries, as well as healthcare professionals, this study is a clear indicator of where the future is headed—towards more personalised, predictive, and preventive care. As the research continues to evolve, it will be fascinating to see how these findings are translated into everyday medical practice, offering new hope for patients worldwide.

Pharma IQ will continue to follow this story and bring you updates as Dr Chris Foley and Dr Riccardo Marioni, along with their teams, continue to push the boundaries of what’s possible in disease prediction and prevention.

Further Reading 

For more insights on how AI is maximising ROI in Drug Discovery and Development, download Pharma IQ's Report. 

Find out how AI is Maximising Patient Recruitment and Retention in Clinical Trials with these Frequently Asked Questions.

Join Us to Learn More

Eager to dive deeper and learn how to overcome integration challenges from real-world case studies?

Join industry leaders and experts at the AI Pharma & Healthcare Event, where you can gain firsthand insights into successful AI integration strategies. This is your opportunity to learn from the successes and challenges faced by others and to network with professionals who are at the forefront of AI technology in the pharmaceutical industry.

đź“…Event Details

Event: AI Pharma & Healthcare Summit
Dates: 24-26 September 2024
Location: Amsterdam Marriott Hotel
Event Website: AI Pharma & Healthcare Summit

Read the full study in Nature Aging:

Gadd, D.A., Hillary, R.F., Kuncheva, Z. et al. Blood protein assessment of leading incident diseases and mortality in the UK Biobank. Nat Aging 4, 939–948 (2024).

Blood protein assessment of leading incident diseases and mortality in the UK Biobank

Received: 15 March 2023
Accepted: 22 May 2024
Published: 10 July 2024


RECOMMENDED