Bits In Bio Weekly - April 29th 2025
Updates From and Around the Bits in Bio Community
📢 Highlights
Repertoire Immune Medicine lands $765 M deal with Genentech for TCR Platform Technology
Eil Lilly and DCVC-backs Grove Pharma’s Polymer-protein undruggable PPI targetingPolymer-protein hybrids with 30M series A
Tempus rolls out ‘Loop,’ a real-world-data-plus-organoid engine through collaboration with Astra Zeneca and Pathos
Special Announcement
Synbiobeta returns to San Jose, CA for its 14th year to the Bay Area, and BiB members get a 200$ discount, just use promo code
BNB-SBBwhen you sign up for a Conference or Exhibit Hall Pass online. You can also click here to download the most recent agenda and check out the programming in the world’s longest running synthetic biology conference.Not yet a member of our super awesome slack community of >8500? Join HERE 🤗
👀 In Case You Missed it …
Repertoire Immune Medicine’s DECODE platform to chart T-cell targets in Genentech autoimmune pact
Genentech is paying up to $765 million with $35M up front and the rest in milestones—to Repertoire Immune Medicine for access to its highly-dimensionalized map of T-cell receptor, antigen, and HLA interactions along with their digital analysis and targeting platforms. The partners will focus on a single, undisclosed autoimmune disease, with Repertoire discovering targets and Genentech engineering therapies. DECODE’s goal is antigen-specific tolerance that spares the rest of the immune system. If successful, it would move immune-modulating drugs from blunt force to precision rewiring. Royalties flow back to Repertoire should any candidates reach market. Based in Cambridge, the world class team has been working diligently with what is rumored to be one of the world’s largest and most diverse immunological datasets, and is adding this win to their announcement last year of a $1.8 billion dollar deal with Bristol Myers Squib. As part of the announcement, Genentech’s immunology head praised the chance to “bring medicines to patients that reset the immune system,” highlighting the potential to truly address root causes rather than just symptoms.
Grove raises $30 million Series A for fully synthetic, cell-permeable biologics.
Chicago-based Grove Biopharma closed a $30M Series A to propel its Bionic Biologics™ platform – a new therapeutic modality that melds biologics and synthetic chemistry to hit intracellular targets previously deemed “undruggable”. Bionic Biologics are essentially fully synthetic, protein-sized polymers designed to act like biologic drugs but with the crucial ability to permeate cells. By integrating precision polymer chemistry with computational and protein engineering tools, Grove can craft multivalent molecules that engage large protein–protein interaction surfaces inside cells (think oncogenic transcription factors or scaffold proteins) that antibodies can’t reach and small molecules can’t easily. The result is an “integrated platform capable of designing fully synthetic, cell-penetrant, protein-scale molecules that solve protein-scale problems” – in the company’s words. Early proof-of-concept is encouraging: Grove’s lead program, an androgen receptor degrader for castration-resistant prostate cancer, has shown in vivo efficacy and is marching toward an IND. The funding (led by DCVC Bio with participation from Eli Lilly and others) will expand the pipeline, initially focusing on oncology and neurodegenerative diseases where intracellular targets abound. If successful, Bionic Biologics could open up a pharmacopeia of new drug targets – essentially bridging the gap between biologics and small molecules.
Tempus AI, AstraZeneca, and Pathos Collaborate to Build Multimodal Oncology Foundation Model
Coming together to form a potent alliance, Tempus AI, AstraZeneca, and Pathos have announced a collaboration to develop a multimodal foundation model for oncology, aiming to accelerate cancer drug discovery and clinical trial development. Tempus will contribute its extensive repository of de-identified oncology data—which now surpasses the Cancer Genome Atlas (TCGA)—including over 8 million patient records, 2 million imaging datasets, 1.5 million matched clinical and genomic profiles, and more than 250,000 transcriptome datasets. The new foundation model will be jointly leveraged by all three organizations to uncover novel drug targets and streamline clinical trial efforts. This partnership also builds on Tempus and AstraZeneca’s existing work in lung cancer, where they are collaborating on biomarker discovery to advance precision diagnostics and treatments.Following Tempus’ recent acquisition of Deep6 AI, a company specializing in AI-driven patient-trial matching, this expanded alliance further strengthens Tempus’ efforts to accelerate patient enrollment and optimize trial development across the oncology space.
Nuclera delivers lightning-fast membrane-protein workflow to its benchtop eProtein printers
UK-based Nuclera has upgraded its eProtein Discovery™ benchtop protein printer with a new workflow for membrane proteins, enabling rapid expression and purification of these notoriously tricky targets in as little as 48 hours. Membrane proteins (think GPCRs, ion channels, transporters) are critical drug targets – ~60% of meds hit them – but their hydrophobic nature makes them hard to produce in traditional cell cultures. Nuclera’s solution: a cell-free expression platform combined with digital microfluidics and an automated additive screen to test solubilization conditions on the fly. In a recent demo, the system synthesized two challenging membrane proteins – MsbA (a bacterial transporter) and ZMPSTE24 (a human membrane enzyme) – and within 24 hours had automatically identified the optimal mix of nanodiscs, lipids, and detergents to keep them happy and folded. By the 48-hour mark, eProtein had scaled up production and yielded large quantities of these proteins that were functionally active and cryo-EM ready. The best part: existing eProtein instrument users get this new capability via a simple software update, instantly adding high-throughput membrane protein expression to their toolkit. This advance could accelerate drug discovery for membrane targets – researchers can go from DNA sequence to purified, structurally-stable membrane protein in 2 days, dramatically shortening a process that used to take weeks or months of trial and error. For pharma and academia wrestling with the membrane protein bottleneck, Nuclera’s all-in-one platform promises to make the “undruggable” a lot more accessible.
Avidicure launches with $50 million to build multifunctional “AVC-Booster” antibodies.
Amsterdam-based Avidicure emerged from stealth with seed financing to develop dual-agonist, avidity-engineered antibodies that activate innate and adaptive immunity only inside tumors. Lead program AVC-S-101 zeroes in on TROP2-positive lung cancers and other solid tumors. Pre-clinical data suggest stronger tumor killing and lower cytokine spillover than first-gen checkpoint blockers or CD3 engagers. Backers include EQT Life Sciences and several European VCs, and the company plans to debut data at AACR and ASCO next year. If the tech works, AVC-Boosters could compress multiple immunotherapy tricks into one well-behaved molecule.
Tempus unveils Loop, an AI platform marrying real-world data to CRISPR-screened organoids.
Released by the biomedical data powerhouse Tempus, Loop clusters millions of cancer patients into molecular subgroups, predicts driver genes with multimodal AI, then knocks those genes out in matched patient-derived organoids. High-throughput CRISPR screens confirm which targets actually matter in human-like tissue, feeding results back into the model for continuous learning. A pharma pilot validated multiple oncology targets in roughly a year—far quicker than the usual three-to-five-year slog. Tempus believes the closed-loop workflow can shave risk off early discovery and shrink the 90 % failure rate haunting pre-clinical assets. Commercial roll-out starts with oncology partners this year.
Duke researchers run a national lung-cancer screening study entirely in silico.
Researchers at Duke University built 294 digital twins with realistic anatomy, imaged them with simulated CT and X-ray, and let deep-learning “radiologists” read every scan. The “virtual trial” reproduced the National Lung Screening Trial’s core finding: CT’s AUC hit 0.92 versus chest X-ray’s 0.72 for nodule detection. Matching real-world specificity within 1 percentage point, the study suggests large screening hypotheses can be tested on cloud GPUs for a fraction of the time and money. Next up: layer in longitudinal growth and survival to mimic full phase-III endpoints. Clinical trials won’t disappear—but they might start on silicon, though its hard to tell how well this method will translate to real world routine care settings.
Axiom nets $15 million seed to predict drug-induced liver injury with AI
Backed by a $15M seed led by Amplify and Zetta Venture Partners. Axiom Bio has set out to eliminate drug toxicity failures before they reach animal or human trials – by using AI models trained on human cell data instead of live animals. Co-founded by Brandon White and Alex Beatson, Axiom spent its first year building what it calls the world’s largest human-first toxicity dataset: over 115,000 small molecules tested on proprietary in vitro human liver models, generating high-content images and multiplexed biochemical readouts for each. They coupled this with a curated clinical outcomes database (including 700+ drugs labeled for liver toxicity risk via FDA’s DILIrank) to teach their AI how in vitro signals translate to real-world drug-induced liver injury. The result is an AI-driven toxicity prediction platform that, in initial reports, can flag hepatotoxic compounds with ~70–75% sensitivity while maintaining ~90% specificity for non-toxic compounds. Impressively, this performance meets or exceeds that of current gold-standard lab assays used by pharma (such as liver spheroid models), and does so at a fraction of the cost – on the order of $100–$450 per compound screened vs. $3k–$15k for traditional wet-lab tests. Axiom’s vision is to integrate such AI models into the drug pipeline, so that early-stage compounds can be vetted for toxicity with human-relevant data and machine learning, before any animal testing or clinical trials occur. If it works, pharma could avoid pushing likely-toxic candidates forward (saving time, money, and animal lives), and regulators might eventually accept in silico/in vitro toxicity data in lieu of some animal studies. It’s early days, but Axiom’s liver model is a compelling proof-of-concept – unveiled publicly at SOT 2025 – hinting that AI plus human-cell assays can predict liver toxicities more predictively than sacrificing more lab rats.
Flagship launches Etiome to forecast—and intercept—disease progression
Spun out of Flagship, Etiome’s Temporal Biodynamics™ platform ingests population health records, single-cell omics, and other data to chart how diseases unfold. By forecasting how individuals will advance along a disease continuum and identifying stage-specific biomarkers (dubbed Biostage Markers), the platform aims to enable Biostaged Medicines – interventions tailored to halt or reverse disease at early, still-manageable stages. The venture spent four long and quiet years in Flagship’s foundry and now emerges with a $50 million internal commitment to build its first pipeline. Initial focus areas include metabolic and neurodegenerative disorders where early intervention could bend cost curves. Think of it as a weather app for your biology—predict the storm, reroute in time.
Synthetic Design Lab debuts with $20M seed funding to super-charge antibody–drug conjugates.
Fueled by a fresh seed round of $20M, Synthetiq has built the SYNTHBODY™ platform to pack at least ten times more cytotoxic payload onto each antibody. Higher drug-to-antibody ratios aim to wring efficacy from tumors that express lower levels of target antigen, widening ADC reach. The $20 million seed round, led by Playground Global, will push a first clinical candidate into trials by 2026. Engineering focus spans Fc geometry, linker chemistry, and AI-guided stability screens. If the physics holds up in humans, SYNTHBODY could make “magic bullet” therapy less of a slogan and more of a statistic.
🗓️ Upcoming BiB Events
See our handy dandy Lu.ma event calendar HERE, please RSVP so folks can plan accordingly!
Minneapolis-St. Paul Get Together, Tuesday, May 6 at 5:00 PM
Thank you to Brian Repko for organizing
SynBioBeta BiB Happy Hour with Trinet, Wednesday, May 7 at 5:00 PM
Thank you to organizers Shantenu Agarwal, Beko Jang, Eti Sinha, and Poonam Lata
Chicago Bits in Bio Happy Hour, Tuesday, May 29 at 6:00 PM
First event in Chicago!
Thank you to Matthew Schechter for organizing
📰 Top Community Conversations
Check out folks suggestions for python packages for publication-ready figures!
🏢 New Job Openings
Part-Time Executive Director at Association of Science Communicators
Mass Spectrometry Software Scientist at Genedata
Senior Computational Structural Biologist at Proteinea
Senior Product Marketing Manager at Definitive Healthcare
🙋 🙏 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!