Recursion Pharmaceuticals

For FAP Awareness Week, we want to highlight one of the strongest advocates in the space.

Living with the rare disease familial adenomatous polyposis (FAP) means a lifetime of uncertainty, surgeries, and complications. Jenny Jones, founder of the Life’s a Polyp Foundation, along with her dad, Timothy Jones, shared what that journey has looked like for their family – where not only Jenny, but her mother and grandfather, struggled with the disease which results in hundreds to thousands of polyps developing in the GI tract that have 100% likelihood of turning cancerous if not removed.

Once she was diagnosed at age 8, Jenny says, FAP became a part of her daily life. By age 10, she’d had multiple surgeries, including the removal of her colon. Her dad called it a miracle that she survived.

The major surgeries continued throughout her life – and have led to numerous challenges, including nutrient deficiencies, dehydration, limitations in what she can eat, painful bowel movements, and medical PTSD. “It’s something that’s always on the mind,” she says.

Jenny has now dedicated her life to serving as a resource and building a trusted network for the FAP community via her foundation, and supports research into new treatment options for FAP patients who currently have no non-surgical options available.

Recursion is working toward solutions for patients with FAP.

👉 Learn more at: https://recursion.com/

FAPawarenessweek

Recursion is advancing the first potential drug for rare disease FAP, discovered with AI.

For FAP Awareness Week, we’re resharing our video featuring CEO and President Najat Khan announcing the first clinical validation of Recursion’s AI-enabled drug discovery platform – positive Phase 1b/2 data for AI-identified drug REC-4881 in Familial Adenomatous Polyposis (FAP).

This progressive rare disease – which often begins in the teens and results in hundreds or thousands of polyps developing in the GI tract which have a 100% likelihood of becoming cancerous if not removed, requires a lifetime of invasive surveillance and life-altering surgeries. There are currently no FDA-approved treatments.

Recursion used AI not only to deliver the first potential treatment for this disease but also to quantify – for the first time at scale – the true natural progression of polyp burden in the trial-relevant population. This program is the first clinical validation of the Recursion OS, and demonstrates our ability to translate unbiased phenotypic insights into potentially differentiated treatments for diseases with high unmet need.

We’re working hard to advance new solutions for FAP patients, this week and every week.

“Just like we simulate molecules and we simulate biology, what if we could simulate our trials before we run them?”

In this clip from Recursion CEO and President Najat Khan’s conversation with Selina Koch for the BioCentury Show, she talks about where AI is making an impact on clinical trials.

💠Recursion’s approach is focused in three areas, Najat says.
1. Improving patient stratification - “Can you improve the signal to noise to know which patients would better respond?”
2. Smarter protocol design and simulating the protocol.
3. Improving recruitment and enrollment. “80% of trials don’t recruit on time.”

👉 Watch the full conversation here: https://www.youtube.com/watch?v=LfGy7FvHk0A

🧬 Bridging the gap in AI drug discovery.

Ali Denton, Staff Machine Learning Scientist at Recursion and one of the authors on the recent paper in Nature Biotechnology, explains how the AI model TxPert predicts how a cell will respond to perturbations.

Predicting a cell’s RNA activity, or transcriptome, is key to bridging the gap between cellular changes and clinical outcomes and advancing the potential for AI drug discovery. As Ali says, “with hundreds of cell types and so much disease variation, the total possibilities are too vast to measure in a lab.”

She describes how TxPert allows us to perform a “Virtual Assay,” taking the mathematical signature of a healthy cell called the Basal State and adding the perturbation’s embedding to deliver a highly accurate prediction of what the cell’s transcriptome will look like after treatment.

TxPert uses layered graph-based models that integrate phenomics — or how a cell looks — and transcriptomics — which genes are expressed — along with massive public biological knowledge resources.

The model can even predict how a perturbation will work in entirely new cell lines it hasn’t seen before as well as accurately forecast the effects of “double perturbations,” consistently identifying “unknown unknowns” that traditional models — and even massive general-purpose AI — often miss.
Ali notes that TxPert is currently predicting genetic perturbations, but more flexible models — including those predicting drug effects — are in the works.

👉 Check out the full paper in Nature Biotech: https://www.nature.comarticles/s41587-026-03113-4

“What’s the one question we obsess over at Recursion? ‘How do we harness the full power of AI to consistently and with urgency create better medicines for patients’?”

In a recap of Recursion’s recent 1Q earnings, CEO and President Najat Khan talks about the tangible evidence that Recursion’s AI platform is delivering. They include:

▪️ REC-1245, a potential first-in-class oncology program where both the biology and molecule were discovered using the Recurions OS. Early clinical data shows a well-tolerated profile with no dose-limiting toxicities and an encouraging PK profile, with more data expected in the second half of this year.

▪️ REC-4881, a potential first-in-disease program for the rare disease FAP that has already shown strong proof of concept with meaningful and durable impact. We’ve now initiated engagement with the FDA on a potential path to registration, with an update expected in the second half of this year.

▪️ REC-4539, a potentially best-in-class LSD1 inhibitor for small cell lung cancer and AML in which the first patient was recently dosed in the Phase 1 trial. “This is a precision-designed molecule,” Najat says, “built to address class-limiting toxicity” with the potential for “improved safety and CNS penetration.”

It’s not about one asset, Najat says, but building a “repeatable, AI-driven product engine that’s starting to deliver across discovery and into the clinic.”

Read Recursion’s full earnings report here: https://ir.recursion.com/news-releases/news-release-details/recursion-reports-first-quarter-financial-results-and-provides

techbio

“Bigger isn’t always better.” — Dave Hallett, CSO of Recursion

At the recent SynBioBeta event, Dave joined fellow industry leaders from Xaira Therapeutics, GSK, NOETIK and NVIDIA to discuss one of the most pressing issues in AI drug discovery: “Solving the Scale Mismatch Between Cells and Patients in Virtual Biology.” How do we better translate cellular insights to patient outcomes using AI – and change drug discovery’s 90% failure rate?

💡 Key insights include:

▪️ Context Is Everything: “It is important to generate high quality, high value, large perturbational datasets, but you also need to do that in context,” Dave said. That’s why Recursion is increasingly moving toward multimodal data generated in highly specific contexts.

▪️ Engineering Needs to Match Human Reality: As Dave pointed out, the closer a model gets to a human, the harder it is to scale perturbational data. Recursion is bridging this gap by moving toward complex, engineered systems like iPSC-derived neurons and engineered cancer cell lines, and then, “Perturbing those at genome scale. And then integrating in over a dozen cellular systems,” merging phenotypic, transcriptomic, and proteomic data.

▪️ Quality Over Scale: Dave noted that while massive public datasets like single-cell transcriptomics exist, the variation in how that data was collected across different labs often results in models that memorize technical noise rather than biological truth. To build accurate foundation models—like Recursion’s recently announced state-of-the-art transcriptomic foundation model known as TxFM—he said we need high-quality, standardized, multimodal data and model architecture that accurately reflect unordered, interconnected biological states.

Thank you to Stacie Calad-Thomson for moderating an excellent discussion, and to Marc Tessier-Lavigne, Kim Branson, and Ron Alfa for sharing their perspectives on the future of virtual biology.

AI

Today, we announce our Q1 2026 business updates and financial results – demonstrating continued momentum across our internal portfolio and partnered programs, with multiple milestones achieved or on track.

🚀 Key proof points include:

▪️ REC-1245 (RBM39 degrader): Early clinical data demonstrate a well-tolerated safety profile and predictable, dose-dependent PK (n=16); dose escalation ongoing with no dose-limiting toxicities observed to date.

▪️ REC-4539 (LSD1 inhibitor): First patient dosed in Phase 1 trial; platform-derived, selective, brain-penetrant profile with a potentially reversible mechanism and shorter predicted half-life aimed at reducing on-target platelet toxicity, supporting differentiation in solid tumors and AML.

▪️ REC-4881 (MEK1/2 inhibitor): Strong Phase 2 efficacy signals and a safety profile consistent with the MEK inhibitor class, with FDA engagement initiated to define a potential registrational pathway and an update expected in 2H26.

▪️ Our joint programs with Sanofi continue advancing towards development candidate designation and earlier stage program milestones in the next 12 months, and we expect to continue to translate biological insights from maps delivered to Roche and Genentech into potential target validation milestones over the next 12 months.

As CEO and President Najat Khan says, this “represents a growing set of proof points that demonstrate our ability to translate platform insights into clinical programs. This progress reflects the strength of both our internal pipeline and partnerships, with multiple differentiated programs advancing through our end-to-end AI platform.”

Our Q1 Earnings report also highlights our disciplined capital ex*****on: reiterating the 2026 guidance of

🧬 Closing the translation gap between cells and patients. 😷

Nature Biotechnology just published a new paper from Recursion on TxPert – a deep learning framework that accurately simulates the transcriptomic shift in unseen biological contexts. TxPert represents an important step in our ongoing work to accurately model transcriptomics and bridge the gap between in vitro discovery and clinical reality – which is critical for improving and scaling AI drug discovery.

👉 Read the full publication in Nature Biotech here: https://www.nature.com/articles/s41587-026-03113-4

We had a fantastic time with Valence Labs at ICLR in Rio sharing our latest machine learning breakthroughs, including presentations on TxFM, our state-of-the-art transcriptomics model that outperforms models up to 100x larger in terms of data size, and MarS-FM, our new class of generative models for molecular dynamics simulations.

And there were lots of great community conversations happening at the rooftop TechBio Social, co-hosted with ICLR’s Learning Meaningful Representations of Life (LMRL) Workshop.

Coming soon: we’re looking forward to sharing more of our ML breakthroughs at !

👉 TxFM paper here: https://openreview.net/pdf?id=NqZqClqtTK

👉 MarS-FM paper here: https://arxiv.org/html/2509.24779v3