Disease prediction and new drugs: why UK Biobank’s huge new protein project matters

Unprecedented amounts of proteomics data will reveal how to predict and treat conditions as diverse as dementia and multiple sclerosis, researchers suggest

“I'm ecstatic,” says health researcher Nophar Geifman about UK Biobank expanding its analysis of proteins – our bodies’ molecular workhorses – from 54,000 to all half-a-million participants. She and others see potential for breakthroughs in detecting and treating dementia, heart disease, depression and many other conditions. 

Proteins’ jobs range from holding together our skin and bones to recognising pathogens. Our bodies contain an enormous number of different proteins: estimates range from tens of thousands to several billion. Proteomics looks at proteins on a large scale, analysing thousands of them at the same time. 

While genes are mostly fixed at conception, our bodies’ protein landscape (aka our proteome) can change dramatically depending on our environment, lifestyle, health and more. Diseases that are affected by things such as “diet, exercise, where we live, the type of work we do, our ability to go see a doctor early – you're going to be able to capture that more in the proteome than you are in the genome”, Geifman explains.

"Proteomics is going to completely revolutionise how we're able to detect those diseases early."

Benjamin Jacobs, neurology researcher

In 2023, UK Biobank’s Pharma Proteomics Project released data from a pilot analysis of 54,000 participants’ blood samples. It was the biggest study of its kind – few proteomic studies had exceeded 10,000 samples until then – and yielded hundreds of research studies within just one year. 

One team revealed certain proteins that serve as early warning signs of 19 cancer types. Another found that small numbers of proteins can predict someone's risk of 67 health conditions better than information clinics typically consider. Recently, researchers discovered that people who experience loneliness have higher levels of proteins linked to heart disease, type 2 diabetes and stroke. 

Ten times more

Now, the Pharma Proteomics Project is set to analyse ten times more blood samples: those from all 500,000 UK Biobank participants, plus 100,000 second samples taken up to 15 years after the first. 

The data will be particularly valuable for understanding conditions that are only partially determined by our genetic background: Alzheimer’s disease, chronic kidney disease, mental health conditions, cardiovascular diseases, lupus and other autoimmune disorders. “Proteomics is going to completely revolutionise how we're able to detect those diseases early,” says neurology researcher Benjamin Jacobs.

Blood tests that look at a person’s proteins could eventually identify health conditions long before they affect someone’s life. “Once you found [a protein] signature that identifies, for example, depression at its early stages, that can now become a fingerprick test that a GP can administer at very low cost to all their patients to try and pick up that signature before even symptoms occur,” Geifman suggests. Some of those efforts are already underway: researchers who found four blood proteins that predict dementia up to 15 years before diagnosis are now working on creating a dementia blood test.

"Because UK Biobank is so huge, and it's linked to real world NHS data and diagnoses, you can study diseases that nobody has paid attention to."

Claudia Langenberg, public health specialist

“One of the most exciting things we would be able to do with 10 times more data is starting to look at protein predictors of how MS [multiple sclerosis] progresses over time,” says Jacobs. Many diseases manifest differently in each person and knowing how the condition might progress could help people live better with it. Public health specialist Claudia Langenberg agrees that using protein tests to predict someone’s prognosis “would have huge potential to be implemented”.

Researchers think that integrating protein data with genetic data will reveal more about diseases’ biological mechanisms, which could lead to new treatments. The biggest value of UK Biobank’s huge amount of data, says Langenberg, is that it allows researchers to study diseases that nobody has paid attention to. “By being agnostic and systematic, those diseases will all get attention.”