We scan new podcasts and send you the top 5 insights daily.
John Baker says IMU has about 35 people in Canary Wharf, the vast majority of its global team. He claims the wet-lab and in-silico platform is efficient enough for that team to handle the full Biobank volume, which runs from a few hundred samples now to tens or hundreds of thousands.
Generating millions of data points for AI requires industrializing lab workflows. To avoid data degradation from cell stress during long experiments, the Xaira team introduced chemical fixation to preserve cell states and re-engineered processes for time-shifted operations, ensuring consistent, high-quality training data.
In a UK-wide immuno-oncology program, John Baker says IMU's pre-treatment immune profile separates patients likely to respond to anti-PD-L1 or checkpoint inhibitor combinations. It also flags before the first dose who may suffer severe immune-mediated adverse events, so clinicians can monitor and manage those patients more actively. Response rates are only 30-40%.
John Baker says the founding team opened a £1 profiling shop on King's College campus in 2021. They used it to build the first high-resolution dataset of 1,000 immune systems, now shown on the office wall as the healthy-immune reference. That scrappy sampling gave IMU a population baseline before UK Biobank.
Challenging the notion that biotech requires large teams for clinical progress, Coursera Health operates with just nine people. This small team is managing one Phase 1 trial while initiating a second and developing an AI platform. This demonstrates how a focused, highly experienced team can achieve major milestones with extreme capital efficiency.
Infinimmune’s platform bypasses traditional discovery methods by reading antibodies directly from human memory B cells. The core insight is that the immune system has already spent a lifetime selecting and validating the most effective antibodies, providing a superior, de-risked starting point for new human therapeutics.
The UK Biobank's decision to allow broad access to its genetic data for both commercial and academic researchers resulted in a 100x greater impact than more restrictive biobanks in the US. This success highlights how open data strategies can dramatically accelerate scientific and commercial innovation.
John Baker says live Biobank blood cells are destroyed once a tube is taken out and run on IMU's platform, and many samples are aged. To maximize each one, IMU extracts hundreds of millions of proteomic immunology measurements plus high-resolution single-cell gene expression, then uses neural networks to map those datasets together.
Unlike purely in-silico companies, Metaphor's platform starts with high-throughput wet lab experiments to generate massive datasets on receptor interactions in living systems. This real-world data is crucial for training their AI to design functionally active antibodies.
To overcome the scaling challenges of traditional biobanks, Regeneron is pioneering a new model. They partner with companies specializing in aggregating de-identified health records and, separately, with groups handling bio-sampling. This "uncoupled" approach allows them to link massive, independent data streams to achieve unprecedented scale.
John Baker says the immune system shifts with breakfast, exercise and circadian rhythms, yet some anchoring features stay consistent for years or decades. IMU wants to separate those stable anchors from the modifiable parts so it can understand disease progression and treatment effects across a population over time.