Creating a Partnership Between the UK Biobank, Philanthropists, and Pharmaceutical Companies
We have written before about the importance of large, open datasets to unlock the benefits of AI in areas like climate, health, and education.
A model example of such a dataset is the UK Biobank. Based in Oxford, the UK Biobank has been regularly collecting health and genetic data from over 500,000 participants across the UK since 2006 – that is almost 20 years!
In 2022, members of our team launched a collaboration with the UK Biobank to scale its infrastructure and do more advanced imaging and genomic sequencing of patients. Ultimately, with help from partners, we were able to raise £32 million in philanthropic commitments, along with an £8 million in-kind contribution of cloud storage from AWS, and further investments from pharmaceutical companies like Regeneron.
In this Q&A, we spoke to Tom Kalil, CEO, and Ronit Kanwar, Head of Strategy and Partnerships at Renaissance about how this collaboration happened and what we can learn from it.
Caption: The UK Biobank recently released the world's largest-by-far single set of sequencing data
Q: Tom, what is the UK Biobank and why is it a big deal?
Tom: The UK Biobank is the world’s most comprehensive dataset of longitudinal health data, with over 15 years of data from over 500,00 participants across the UK on:
Lifestyle
Health information
Proteins found in the blood
Whole body imaging scans
The UK Biobank is important because we still do not understand the genetic and biological basis of most diseases. Researchers can use data from the UK Biobank to better predict, diagnose, and treat diseases like cancer and Parkinson’s, in the same way that Google DeepMind was able to use the Protein Data Bank, a large open dataset of protein structures, to train AlphaFold, an AI model that can predict protein structure based on a sequence of amino acids.
Q: Tell us about how this collaboration happened. How did it start? How did you succeed?
Ronit: In 2022, members of our team started speaking with the leadership of the UK Biobank. We learned that, while the UK government was providing significant support, there were certain kinds of advanced genomic and imaging data which were not being gathered. In addition, the UK Biobank was not fully integrating data from the UK’s National Health Service (NHS). While they were not initially thinking about approaching philanthropists to help fill these gaps, when our team suggested it they were open to it.
At the same time, we started approaching philanthropists to gauge their interest. Some of them were interested in accelerating biomedical progress broadly. Others were interested in specific diseases like Parkinsons. It took time to explain why co-funding a large government-backed project was a catalytic use of funding, but we were able to address this question.
Tom: One way we did so is by explaining to them the potential to use the UK Biobank to identify biomarkers for diseases that they care about.
The societal value of biomarkers is huge! Economist Heidi Williams has estimated the social Net Present Value of better biomarkers for cancer at $2.2 trillion! That’s because biomarkers that the FDA has confidence in can (a) shorten the duration of a clinical trial; (b) make it more attractive for companies to invest in drug discovery; and (c) increase the rate at which survival time for cancer patients improves.
One way to discover and validate biomarkers is to do large-scale proteomic studies. As the cost of proteomics has dropped, it is now possible to contemplate doing a proteomic analysis of all 500,000 UK Biobank volunteers A pilot project analyzed the abundance of nearly 3,000 proteins in blood samples in 54,000 volunteers. This study revealed, for example, that inflammatory proteins are significantly higher in patients with depression, which will be useful for drug discovery efforts.
In 2023, philanthropists committed £32 million to the UK Biobank which was matched by the UK government. In 2024, we’ve seen additional commitments of £8 million in cloud storage credits from AWS, and further investments from pharmaceutical companies like Regeneron in specific genomic sequencing projects. But we’re still short of the potential.
Q: What should we take away from this?
Ronit: Our work with the UK Biobank shows how philanthropists, companies, and public actors can work together to scale a critical public good. We hope to continue to forge multi-sector partnerships tackling important challenges.
Another example of how we’d like to promote multi-sector partnerships is increasing the use of market shaping tools like advance market commitments (AMCs) to create demand for a product with high social value but low commercial value.
For example, in 2022, Stripe, Meta, Alphabet, Shopify, and McKinsey launched Frontier Climate, a $925 million AMC for carbon removal.
We are interested in working with companies, philanthropists and governments to identify, design and launch the next 3-5 market-shaping initiatives in sectors that contribute 0.5 Gt or more in CO2e. This could involve:
CEO-level outreach to groups such as the Business Roundtable and WEF to identify sectors where companies are willing to make significant pledges. This would build on previous efforts like the First Movers Coalition – but would push for more specific commitments.
Working with groups like UChicago’s Market Shaping Accelerator to develop a market-shaping approach (e.g. AMCs, milestone payments, volume guarantees, off-take agreements, contract for differences, pooled purchasing by private sector) for a given sector that is likely to have a significant impact.
Examples of sectors to focus on:
Industrial heat
Alternative proteins
Geothermal
Cement
Enteric methane (e.g. vaccines)
Carbon-neutral replacement for palm oil
Green ammonia
