Bits In Bio Weekly - May 20th 2025

Updates From and Around the Bits in Bio Community

📢 Highlights

• CRISPR-for-1, Philadelphia Doctors Deploy Bespoke Gene Edit to Treat a Sick Infant

• Aussie Stem Cell Breakthrough Nets $35M Partnership with Retro Biosciences

• Molecular Glue Goes Prime Time—Blueprint and VantAI ink AI-first megadeal

• Regeneron Buys Assets of 23&me Out Of Bankruptcy for $256M, Ending Dramatic d2c Testing Saga

  Not yet a member of our super awesome slack community of >9000? Join HERE 🤗

---

👀 In Case You Missed it …

A one-of-a-kind CRISPR therapy saves infant, hinting at an ultra-personalized future for genetic medicine

In Philadelphia, a 10-month-old boy became the first patient to be treated with a customized CRISPR gene therapy made just for him. Born with a deadly metabolic disorder that prevented him from processing proteins, baby KJ faced a dire prognosis – until doctors at Children’s Hospital of Philadelphia (CHOP) and Penn Medicine engineered a bespoke genetic fix targeting his exact mutation. In an unprecedented sprint, the team designed and manufactured the CRISPR-based therapy in about six months and administered it in three doses. Now, remarkably, KJ is alive and doing well, though it’s too early to declare him cured. The case, published in NEJM, proves that “N-of-1” medicine is possible: using gene-editing tools to craft one-off treatments tailored to a single person’s genome. It builds on the idea of individualized therapeutics (recalling a 2018 case where a custom RNA drug saved a girl with Batten disease), taking personalization to new heights. This breakthrough could open the door for other ultra-rare disease patients who currently have no options – though it raises big questions about how to scale and pay for such one-and-done cures. For now, it’s a stunning glimpse of the future: a world where even a one-in-a-million genetic mutation might be corrected if you have the right team and a cutting-edge toolkit at the ready.

Lab-grown blood stem cells get a $35M shot in the arm from an Aussie–Silicon Valley team

Australia’s Murdoch Children’s Research Institute (MCRI) has struck a deal worth over $35 million with California-based Retro Biosciences to push lab-grown blood stem cells toward the clinic. MCRI made headlines last year by creating human blood stem cells from ordinary adult cells – a world-first breakthrough that hints patients could “donate” bone marrow to themselves in the future. Now Retro is licensing that tech and funding the next steps to turn it into personalized therapies for bone marrow failure, leukemia and beyond. The idea is to eliminate the need for donor transplants (and the nasty immune rejection risks) by reprogramming a patient’s own cells into brand-new blood stem cells. Retro, a well-funded longevity startup, sees this as a way to extend healthy lifespan by rejuvenating the blood system – a moonshot that aligns with its mission to add years to life. MCRI hopes to get first human trials going within five years, showing how fast a lab discovery can move toward saving lives when academia and industry join forces. It’s a notable cross-continental partnership spotlighting Melbourne’s strength in stem cell science and the growing appetite to invest in regenerative medicine breakthroughs.

Behind the Deal: Blueprint and VantAI’s $1.67B Bet on AI to Crack the Molecular Glue Code

Blueprint Medicines has expanded its strategic collaboration with VantAI in a deal now valued at an impressive $1.67 billion, underscoring the biotech industry's rising confidence in AI-driven drug discovery. Central to this partnership is VantAI’s Neo-1 platform, which uniquely combines atomic-level protein structure prediction and generative molecular design to discover molecular glue therapeutics. These novel small molecules have the potential to modulate protein interactions, addressing previously undruggable targets.

The expanded agreement comes as interest in molecular glue therapies reaches a fever pitch. Major pharmaceutical companies like Pfizer and Novartis are also investing significantly, recognizing their versatility across multiple therapeutic areas, particularly oncology. Blueprint’s increased investment follows its earlier $632 million collaboration with Roivant from 2022, VantAI's parent company, indicating sustained confidence in the potential of VantAI’s innovative technology. What sets Neo-1 apart is its ability to generate and evaluate 3D molecular structures de novo, dramatically accelerating the identification of viable drug candidates compared to traditional methods. The deal’s structure, with substantial milestones tied to R&D and commercial success, reflects both the promise and the risk of this cutting-edge approach. By leveraging advanced AI, Blueprint and VantAI are poised to redefine drug discovery and potentially deliver transformative therapies for some of medicine’s toughest challenges.

Regeneron buys 23andMe’s assets, betting $256M on consumer gene arrays for drug discovery

Following a tumultuous bidding process, Regeneron is picking up the pieces of 23andMe, agreeing to buy virtually all of the consumer DNA testing company’s assets out of bankruptcy for $256 million. The deal would fold what is left of 23andMe’s core business – its spit-kit genetic testing service, health reporting platform, and massive biobank of DNA data – into Regeneron, while excluding a side venture (the Lemonaid Health telemedicine unit) that 23andMe had acquired earlier. If approved, what remains of 23andMe will exist as a wholly owned Regeneron subsidiary, so customers might not notice much change on the surface. But behind the scenes, Regeneron plans to leverage the companies trove of genetic data for its drug discovery programs, marrying the direct-to-consumer genomics empire with the pharma giant’s genetics-driven R&D (Regeneron’s own Genetics Center has already sequenced millions of human exomes). It’s an unusual but logical match: a pharma company hungry for data is rescuing a faltering DTC genomics pioneer whose value lies in ~14 million saliva samples worth of DNA. Do not be fooled - it is the SAMPLES that are truly of value here, NOT the data itself. Regeneron is treading carefully on privacy, pledging to honor 23andMe’s existing consent and privacy policies and even appointing a special privacy ombudsman to reassure customers their genetic info won’t be misused. While there is no electronic health record data attached to the samples in 23&me’s silos, Regeneron will likely exercise the ‘re-contact’ privileges allowed under the 23&me’s terms and conditions that allow it to re-approach users, something that wasn't feasible until 2022. All this will be quite expensive, meaning that the net cost to turn 23&me’s samples and data into something useful will likely exceed $500M due to sequencing and operational costs. The move caps a dramatic turn for 23andMe – once a SPAC-era high-flyer valued at $3.5 billion, now bankrupt and being salvaged via fire sale – and it underscores that while the consumer DNA testing model struggled to make money, the data itself remains a goldmine for drug hunters.

AlphaFold 3 Guides Bridge Recombinases to Megabase-Scale DNA Rearrangements
New research from the Arc Institute introduces "IS622," a bridge recombinase system that precisely rearranges human DNA at the megabase scale. The team leveraged AlphaFold 3 to model protein structures and inform their engineering approach, helping achieve scarless insertions with up to 20% efficiency and specificity as high as 82%. Their computational methods, including deep mutational scanning of the recombinase protein, enabled the creation of an enhanced system capable of inverting nearly a megabase of DNA. The researchers engineered both the bridge RNA and recombinase components, identifying key mutations like S30T, P54Q, and S243H that significantly boosted activity. In validation experiments, they demonstrated therapeutic proof-of-concept by excising the BCL11A enhancer region implicated in sickle cell anemia. The paper highlights potential applications in treating genetic diseases, including excising disease-causing repeat sequences in Friedreich's ataxia. It also points to exciting synergies with AI tools like Evo for generating DNA sequences. This work showcases how AlphaFold 3 is transforming protein engineering and expanding our genetic manipulation toolkit for TechBio companies developing next-generation therapeutics.

Harvard Loses 100’s of M of Research Grants after Threats from Trump Administration

The Trump Administration has followed up on its threats to Harvard, and has cut $450m in grants. It seems that universities tried to soften the blow by issuing bonds, but when they will come and how it will affect researchers affected by these cuts, as well as their employees (often on a work visa), is still up in the air. Researchers relying on non-federal fundings are also starting to feel the pain, as HHMI cuts its Hannah Gray fellowship. Private philanthropies could have been expected to be a safe haven from the current rampage, but it seems they decided to tow the line, as they may fear to get their tax-exempt status revoked.

Eligo’s bacteria-fueled skin therapy gets a boost from France’s biotech fund

Paris-based Eligo Bioscience has been awarded a $5 million French government grant to advance its wild new approach to treating skin diseases. The startup is programming skin-resident bacteria to produce therapeutic proteins on-site – essentially turning your friendly microbes into tiny drug factories. The funding (part of France 2030 via Bpifrance) will help scale up Eligo’s gene-delivery platform as it heads toward late-stage trials in acne and other inflammatory skin conditions. Instead of slathering on creams or taking pills, Eligo uses a topical nanoparticle to slip DNA into hair follicle bacteria, prompting them to pump out medicine right where it’s needed. It’s a highly localized twist on gene therapy and microbiome medicine, aiming to treat conditions like acne vulgaris with fewer systemic side effects. The $5M boost, on top of the company’s recent $30M series B extension and a Nature-published tech breakthrough, shows growing confidence that a little genetic tweaking of your skin flora could go a long way in dermatology.

Datavant bets on real-world evidence, snapping up Aetion to unite data and analytics

Health-tech player Datavant is acquiring Aetion in a bid to create a one-stop shop for real-world health data and analytics. Datavant’s platform already links a vast network of healthcare data – from 70,000+ clinics’ EHRs to pharmacy and lab records – and now it wants to add Aetion’s expertise in crunching that data to generate real-world evidence for drug development and outcomes research. The idea is to offer life sciences companies an end-to-end RWE solution, letting them seamlessly connect and analyze patient data to answer research questions faster (and maybe cheaper) than traditional studies. It’s the latest move in Datavant’s expansion: the company merged with Ciox Health in 2021 in a $7 billion deal and bought Apixio last year, riding the wave of demand for health data in both clinical and value-based care. Grabbing Aetion is a strong vote of confidence in the value of real-world evidence, which regulators and pharma are increasingly embracing – but it also shifts Datavant from a neutral data broker to an active analytics competitor in the RWE space, potentially unnerving some partners. Company execs say Datavant’s revenue will top $1 billion in 2025, underscoring how lucrative and central real-world data has become as healthcare gets ever more digital. In short, the real-world data gold rush is on, and Datavant just staked a bigger claim.

Viz.ai teams with Sanofi and Regeneron to turbocharge COPD care with an AI co-pilot

SF-based Viz.ai is teaming up with Sanofi and Regeneron to innovate on COPD (chronic obstructive pulmonary disease) treatment, leveraging a healthy dose of artificial intelligence. The multi-year collaboration will deploy Viz.ai’s new AI-powered “COPD module” in hospitals, where it will sift through electronic health records to flag high-risk COPD patients who might be slipping through the cracks. By using natural language processing on doctors’ notes and other data, the software can identify patients in need of better management or specialty care and nudge care teams to intervene sooner. COPD – a progressive lung disease affecting an estimated 392 million people globally – often goes underdiagnosed and undertreated, so the hope is that AI-driven early identification and care coordination can improve patient outcomes and reduce hospitalizations. For Sanofi and Regeneron, which are backing the project, it’s a way to enhance disease management in a population that might benefit from advanced therapies (including a new biologic they’ve been co-developing for COPD). This partnership highlights pharma’s growing interest in digital health tools – using tech not just to find more patients for their drugs, but to genuinely streamline care for diseases that have long been overlooked. If successful, a lot more COPD patients could get the right treatment at the right time, courtesy of an AI assist.

Schrödinger trims staff as even AI-driven drug hunters face the biotech downturn’s sting

One of the most respected computational drug discovery companies, Schrödinger, is laying off about 60 employees – roughly 7% of its workforce – to curb costs and conserve cash. The move is intended to save around $30 million per year and comes “in the face of uncertain times and challenging economic conditions,” as the company put it in a candid LinkedIn post. Unlike some biotech restructurings, these cuts are spread pretty evenly across teams, suggesting Schrödinger is trimming fat rather than axing any single R&D program or pivoting strategy. It’s a bit of a reality check for a company that, on paper, has a lot going for it: a widely used physics-based drug design software platform bringing in revenue, three internal oncology drug candidates in Phase 1 trials, and even a $150 million collaboration with Novartis inked last year. But even biotech darlings with AI cred aren’t immune to the current market slump – investor sentiment has soured and every player is extending their runway. Schrödinger insists its long-term priorities remain unchanged (just executed more efficiently), but the takeaway is clear: the 2025 biotech downturn is forcing tough belt-tightening across the board, and not even an industry darling can escape the cold equations of cash burn.

Generative AI meets the clinic as Absci’s first designed antibody begins human trials

Absci has officially moved from computer simulations to human trials, dosing the first volunteers with an antibody that the company claims that it’s AI platform designed from scratch. The drug, ABS-101, is an experimental treatment for inflammatory bowel disease that zeroes in on TL1A – a hot IBD target that recently inspired Merck’s $10.8 billion acquisition of Prometheus. What’s special is how ABS-101 was created: Absci used generative AI and synthetic biology to conjure up and optimize the antibody in silico, aiming for a molecule that’s highly potent, less likely to trigger immune reactions, and long-lasting enough to be dosed just four times a year. Now a Phase 1 trial (in ~40 healthy volunteers) will see if that design performs as advertised in real life, with initial safety and pharmacokinetic data expected by late 2025. It’s the company’s first AI-designed drug to reach humans – a big moment for the AI drug discovery field, which is eager to prove its algorithms can deliver actual medicines, not just paper predictions. Absci is already cooking up more AI-crafted candidates (including an alopecia treatment slated for 2026 trials), betting that its rapid design platform can outpace traditional R&D and give it an edge in crowded therapeutic areas. All eyes in pharma will be watching to see if a molecule dreamed up by a computer can go toe-to-toe with those invented the old-fashioned way – and perhaps even come out on top.

🗓️ Upcoming BiB Events

See our handy dandy Lu.ma event calendar HERE, please RSVP so folks can plan accordingly!

• Bits in Bio April Journal Club – Cell and Gene Therapies, Wednesday May 21st

• BiB x Nucleate Paris, Tuesday, May 27th

• Chicago Bits in Bio Happy Hour! Thursday, May 29 at Recess in Chicago

• Personalized Medicine: The Role of Biotech in Tailoring Healthcare, Thursday May 29th in Lisbon

• Fireside Chat & Happy Hour: Cloud Infrastructure for TechBio Startups, Thursday, May 29th in San Diego

📰 Top Community Conversations

Not much to report this week - have a question?

Post it in our #ask-bib slack channel

🏢 New Job Openings

• Engineer at Biographica

• Chief of Staff at E11 Bio

• Senior/Lead Software Engineer at Foray Bioscience

• AI Scientist and Data Engineers at Howard Hughes Medical Institute

• Data Solutions Specialist at Ontoforce

• Various Roles at Prima Mente

• Various Roles at Chan Zuckerberg Initiative

• Founding Staff Engineer at Axia Medicine

🙋 🙏 Community Asks

Feedback: How is the Newsletter doing? We’re trying different formats/content. In case the hyperlink above didn’t get your attention, maybe a bright orange button will!

Volunteer: Want to get involved with Bits in Bio, meet new members across the community, and learn about the ecosystem? We are looking for volunteers to help us create great content and manage the community.

---

🙏 Thank you for being a BiB Weekly reader!

We want to deliver what matters most in Bio AI and would love your feedback on how we can do better. Please weigh in as anon here or DM me directly!

Vince Alessi

BiB Editor in Chief / Benevolent Dictator of the Newsletter

---