PETRI DISH PERSPECTIVES

Burma | Chicago
Scientist 🥼🧪🔬
Covering biotech, pharma, and healthcare companies shaping the world 🧬🏥🌎

Stream new episodes every THURSDAY!

https://linktr.ee/maneadkhin

#biotech #pharma #healthcare #podcast #petridishperspectives

All Episodes

PETRI DISH PERSPECTIVES

Episode 32: Recursion

December 04, 2025 • Manead Khin • Season 1 • Episode 32

Send us a text

Can a machine discover a cure faster than a scientist? Welcome to the revolutionary world of Recursion Pharmaceuticals, the pioneering TechBio company determined to solve biology using AI and industrial automation.

In this deep-dive, we explore how Recursion replaced slow, human-biased research with the Recursion OS, a closed-loop platform that generates massive, proprietary biological data. Learn about the company's bold vision, from the crucial $50 million NVIDIA investment and the strategic merger with Exscientia to the immense computational power of the Boltz-2 supercomputer. We analyze their pipeline progress in oncology and rare diseases, discuss the challenges of high cash burn (like the recent pipeline streamlining in November 2025), and examine their critical partnerships with giants like Roche/Genentech. Discover why Recursion is leading the charge to build the digital "Map of Biology and Chemistry" to cut drug discovery time from a decade to mere years.

🎧 Listen now, stay curious, and don’t forget to subscribe for new episodes every Thursday!

https://linktr.ee/maneadkhin

#Recursion #TechBio #AIDrugDiscovery #NVIDIA #Biotech #PharmaInnovation #MachineLearning #OmicsData #Genomics #RecursionOS #PetriDishPerspectives #BiotechPodcast

Support the show

Hello and welcome to Petri Dish Perspectives, the podcast where we geek out about science and the companies shaping the future of healthcare. I’m your host, Manead, and I’m a PhD scientist by training, biotech storyteller by choice. With every new episode released on Thursday, my goal is to deliver digestible pieces of information on healthcare companies under 30 mins.

Well, why listen to this episode? Let me tell you. Well, first of all there was a huge interest in the Tempus AI episode so if you enjoyed that one, this episode will be a treat. Another reason why is the process of drug discovery as we all know is famously limited. It takes over a decade, costs billions, and still fails more than 90% of the time. But what if we could replace the slow, human-biased trial-and-error with a massive, automated system, one that generates a digital map of human biology and lets Artificial Intelligence find the hidden cures?

Today, I wanna talk about Recursion Pharmaceuticals, the company built on the radical idea that we can industrialize drug discovery by merging biology, automation, and colossal computing power. This is the story of a company decoding biology to radically improve lives.

Quick disclaimer, I give full credit to the original articles cited in the references in the transcript!

Grab a coffee or tea, settle in, and let’s jump in!

The Origin Story and the Phenomic Shift

The journey of Recursion Pharmaceuticals begins not in a traditional laboratory, but with a fundamental shift in perspective at the University of Utah around 2013.

Co-founder and former CEO, Dr. Chris Gibson, was a medical and PhD student studying a rare hereditary stroke syndrome called Cerebral Cavernous Malformation (CCM). CCM is a type of cerebrovascular disorder characterized by a collection of abnormally formed, enlarged, and irregular blood vessels (capillaries) primarily found in the brain or spinal cord. These clusters of abnormal capillaries are sometimes called cavernomas or cavernous angiomas and have thin, weak walls, making them prone to leaking blood into the surrounding nervous tissue.

Like many researchers, he was manually screening existing drug compounds, looking for one that might correct the cellular dysfunction caused by the disease. This was tedious, manual work, prone to human error and limited by human capacity.

The turning point came when Gibson began using a computer program to analyze high-resolution microscopy images of the diseased cells. The software wasn't just counting cells; it was quantifying the cellular phenotype, the subtle size, shape, location, and texture changes that form a disease's visual fingerprint.

The computer started identifying promising drug candidates based on recognizing patterns of cellular correction, a far more robust and unbiased method than a researcher’s eye. This was the birth of phenomic screening.

Gibson, along with co-founders Blake Borgeson and Dean Li, realized the power of this approach. Why study one disease at a time when you could study thousands? They made the bold decision to scale this technology to industrial proportions, establishing the core mission: to decode biology to radically improve lives by eliminating human bias.

The Recursion Operating System (Recursion OS): A Digital Foundry

The entire foundation of Recursion is the proprietary, end-to-end platform known as the Recursion Operating System (Recursion OS). This system is a closed-loop engine that systematically generates, processes, and learns from biological data.

The Industrial Wet Lab: Data Generation at Scale

Recursion's labs in Salt Lake City are not typical research facilities; they are highly automated factories. The company recognized that public, unstructured biological data is often messy and insufficient for training AI. Therefore, they decided to make their own fit-for-purpose data.

Systematic Perturbation: They start by systematically manipulating human cells (or organoids). This involves techniques like using CRISPR-Cas9 gene editing to knock out genes, chemically inducing diseases, or exposing cells to hundreds of thousands of compounds. They effectively create thousands of in vitro disease models.
Robotic Execution: This process is carried out by industrial robotics, enabling them to run up to 2.2 million experiments per week. This staggering scale ensures comprehensive coverage.
High-Dimensional Data: Each experiment generates millions of microscopy images, capturing the phenotypic response of the cells. This proprietary image and omics data forms a massive phenomic dataset, now exceeding 65 petabytes, one of the largest and most consistently structured biological and chemical datasets in the world.

The AI Brain: Mapping the Unknown

This flood of proprietary data feeds the second, most critical component: the Machine Learning and Foundation Models.

The AI analyzes the phenomic data to understand the causal relationships between a gene (the target), the cellular disease fingerprint (the phenotype), and the compound (the therapeutic correction).

This process builds what Recursion calls the "Map of Biology and Chemistry." This map is a massive, navigable network where the AI can predict:

Novel targets for diseases with no known treatment.
New uses for existing drugs (repurposing).
The optimal molecular structure needed to correct a specific cellular state.

To handle this immense computational burden, Recursion partnered with NVIDIA to build BioHive-2, one of the most powerful supercomputers dedicated to the pharmaceutical industry. The $50 million NVIDIA investment accelerates the training of their foundation models, such as Boltz-2, solidifying Recursion's position at the leading edge of Generative AI for drug discovery.

Consolidation and Pipeline Progress

The journey of a true disruptor is rarely smooth. Recursion has faced intense scrutiny over its high operational burn rate and the pressure to transition platform success into clinical validation.

The Exscientia Merger (November 2024)

In a major strategic move in November 2024, Recursion completed a business combination with Exscientia, a UK-based leader in AI-driven precision chemistry.

Rationale: The combination was designed to create a fully vertically integrated platform. Recursion's strength was target identification and large-scale biological mapping. Exscientia's strength was precision chemical design and rapid de novo molecule synthesis. The unified Recursion OS 2.0 aims to seamlessly move from identifying what to target to efficiently designing how to target it, dramatically shortening the design-make-test-learn cycle.
Scale: The combined company boasted a pipeline of over 10 clinical and preclinical programs and an astounding potential of over $20 billion in milestone payments from partnerships with giants like Roche/Genentech and Sanofi.

The Pipeline and Clinical Validation

The platform is designed to find first-in-class treatments, and it is actively fueling clinical progress:

REC-617: This is an oral CDK7 inhibitor for advanced solid tumors. CDK7 is a protein enzyme that plays a critical role as a master regulator of two fundamental cellular processes: cell cycle progression and gene transcription. It functions as a subunit of the CDK-activating kinase (CAK) complex, which activates other CDKs necessary to move the cell through its division cycle. However, it is difficult to target due to two reasons: dual master regulatory role in the cell and the structural similarity it shares with other proteins, leading to toxicity. Recursion’s design is engineered for a precision profile to manage toxicities. Recent data (Q3 2025) from the Phase 1/2 trial established the maximum tolerated dose (MTD) and showed promising preliminary anti-tumor activity, with expansion cohorts initiated in ovarian cancer.
REC-4881: This drug, which targets a rare disease called Familial Adenomatous Polyposis (FAP), has received Fast Track and Orphan Drug designations from the FDA, underscoring its potential in a high-unmet-need area. FAP, by the way, is a rare, inherited disorder characterized by the development of hundreds to thousands of adenomatous polyps (precancerous growths) throughout the colon and rectum. If left untreated, these polyps have a high risk of developing into colorectal cancer typically by the age of 40. The condition is caused by a mutation in the APC gene.

Recent Headwinds (November 2025 News)

Despite the technological excitement, Recursion's transition from R&D platform to commercial clinical entity remains costly:

Financial Misses: In November 2025, the company reported third-quarter revenue of only $5.2 million, far below analyst expectations. This significant revenue shortfall highlighted the high operational costs and large cash burn associated with maintaining the Recursion OS and industrializing drug discovery.
Pipeline Streamlining: Following the Exscientia merger and disappointing data from some programs (like REC-994 in CCM, which failed to show compelling functional outcomes), Recursion announced a necessary pipeline streamlining to extend its cash runway into late 2027. This involved halting certain development programs to focus capital on the six highest-priority programs with the strongest clinical potential. This is a common and often essential move for TechBio companies, pivoting from platform validation to clinical execution.
Recursion Pharmaceuticals went public on April 16, 2021, in an upsized initial public offering (IPO) that raised $436.4 million. The company priced its Class A common stock at $18.00 per share and began trading on the Nasdaq Global Select Market under the ticker symbol "RXRX". The IPO was a success, with shares closing at $31.30 on the first day, a 74% increase, and the company's market value exceeding $5 billion.

The Strategic Future: Tech Platforms and Partnerships

Recursion's future relies heavily on two things: maximizing its core AI assets and deepening its blockbuster collaborations.

The Power of Partnerships

Recursion's partnership model—working with companies like Roche/Genentech, Sanofi, and Bayer, provides crucial, non-dilutive capital and real-world validation.

Risk Sharing: These partnerships allow Recursion to share the high risk of early-stage discovery while maintaining eligibility for billions in milestone payments and royalties.
Neuroscience Leadership: A key recent milestone was achieving the second neuroscience phenomap milestone with Roche/Genentech in October 2025, specifically the first-of-its-kind Microglia Map. Microglia are the immune cells of the brain, and mapping their behavior could unlock targets for Alzheimer's and Parkinson's diseases.
Recursion Pharmaceuticals has attracted a strong mix of institutional and strategic investors, reflecting the company's unique blend of technology and biology. Its largest institutional shareholders include major index and growth funds like Vanguard Group Inc., BlackRock, Inc., and ARK Investment Management LLC, particularly through Cathie Wood's ARKK Innovation and Genomic Revolution ETFs. Key strategic investors include SoftBank Group Corp., the Abu Dhabi sovereign wealth fund Mubadala Investment Co PJSC, and perhaps most notably, NVIDIA, which made a $50 million investment to accelerate Recursion's AI foundation models for drug discovery. Other significant backers include Baillie Gifford & Co. and Kinnevik AB.

The Vertical Integration Vision

Ultimately, Recursion aims to be the infrastructure layer for drug discovery. By integrating the target finding of the original Recursion OS with the precision molecule design inherited from Exscientia, they are building a "full-stack" system capable of accelerating the discovery process from 10 years to less than 4 years.

The company embodies the bold, sometimes risky, future of medicine. It’s a vision where human expertise is augmented by algorithmic brilliance, and the traditional failures of R&D are overcome by the power of data and automation. Recursion is actively building the infrastructure to make biology "solvable"—a map to guide us to the next generation of life-changing treatments, faster than ever before.

References

Manead Khin

Host