GEN – Genetic Engineering & Biotechnology News
Xaira’s First Virtual Cell Model Is Largest To-Date, Toward Complex Biology
Billion dollar-backed AI drug developer, Xaira Therapeutics, has recently released the largest virtual cell model to-date to predict how cells respond to genetic perturbations in unseen biological contexts. The team asserts that accurately predicting transcriptome-level effects is a powerful translational tool across target and mechanism-of-action discovery, patient stratification, toxicity prediction, and more.
Named X-Cell, Xaira’s model sizes up to a whopping 4.9 billion parameters and has broken barriers as the first scaling law demonstrator in the virtual cell domain. Results showed that perturbation prediction follows power-law scaling with an exponent matching large language models.

In performance, X-Cell achieved zero-shot prediction of T cell inactivating perturbations and effectively generalizes to therapeutically relevant contexts not seen in the training data, including iPSC-derived melanocyte progenitors and primary human T cells from multiple donors. The work is posted as a preprint on bioRxiv that has not yet been peer reviewed.
Xaira launched in 2024 and is led by CEO Marc Tessier-Lavigne, PhD, former president of Stanford and CSO of Genentech. Among the company’s star-studded leadership team are Nobel laureates David Baker, PhD and Carolyn Bertozzi, PhD, former FDA head, Scott Gottlieb, MD, and former CEO of Johnson & Johnson, Alex Gorsky.
Diffusion evolution
Many virtual cell models are primarily fueled by observational single-cell RNA-seq expression datasets. Yet, predicting how cells respond to stimuli, such as treatment with a drug, requires large-scale perturbation data that are sparse in the public domain.
To train X-Cell, Xaira has spent its initial years building what the company describes as “the largest genome-wide CRISPRi Perturb-seq dataset ever reported.” Named X-Atlas/Pisces, the dataset is composed of 25.6 million cells across seven screens and 16 biological contexts, expanding upon X-Atlas/Orion, which was released last June. This unprecedented AI-ready, context-rich dataset enables X-Cell to achieve a parameter scale in the billions.
X-Cell is the first virtual cell model to systematically integrate biological prior knowledge from the literature, such as information on specific genes, protein–protein interactions, and cellular morphology, using a cross-attention mechanism.
In architecture, Xaira’s first virtual cell is a diffusion language model that iteratively refines its predictions by replacing control gene expression values with perturbed values. The method contrasts with the autoregressive training approach, used by previous generation models, including scGPT for single-cell multiomics, developed by Bo Wang, PhD, senior vice president and head of biomedical AI at Xaira.
As an analogy, Wang explained that autoregressive training is like typing from left to right. If the model makes a mistake, the rest of the sentence can fall apart.

In contrast, diffusion language models function like editing a draft. The model starts with a foundation, such as “I like coffee.” The sentence is iteratively refined to “I like decaf coffee,” then “I like finely ground decaf coffee.” With each pass, the model adjusts the output to better match the underlying data distribution.
“It’s more sophisticated and better at predictions,” Wang told GEN Edge when describing the diffusion-based approach. “Experts from the language-side even find it’s better at reasoning.”
Engineering discipline
“The best measure of success is to show that our models can be applied to make medicines and a difference for patients,” said Tessier-Lavigne in an interview with GEN Edge.
He attributes the trial-error nature of science as the culprit for the long timelines and high attrition of drug discovery. Target identification to drug approval takes an average of thirteen years while 90% of molecules fail at the clinic. Xaira’s mission is to transform discovery and development from an artisanal state to an engineering discipline by serving as a platform and product company for the field.
In addition to building the virtual cell, Xaira has a molecular design pillar based on protein design technology developed by Baker who won the 2024 Nobel Prize in Chemistry. The team is developing novel antibodies that hit challenging targets, such as multi-pass membrane proteins with minimal accessible regions outside of the cell. These proteins are therapeutically validated but remain undruggable.
Last November, a Nature study led by Xaira co-founders and Baker lab graduates Nathaniel Bennett, PhD, and Joseph Watson, PhD, generated full length antibodies from scratch that successfully bound user-specified epitopes with atomic precision. In the increasingly crowded de novo antibody space, corresponding advances were demonstrated by Nabla Bio, Chai Discovery, and Absci.
While Xaira has remained quieter on its molecular design efforts, Tessier-Lavigne asserts that the company has been “very focused on this from Day 1.” More public announcements are anticipated in the coming months.
“We’re obsessed with cells”
While virtual cell models that generalize to new contexts provide a valued advance toward understanding biology, predicting patient outcomes is still a step away.
Ron Alfa, MD, PhD, CEO of Noetik, says building up layers of cell-level experiments to achieve a tissue, and ultimately human representation, is a challenging feat. He argues that developing models that tokenize tissues is a more convincing translational approach.
“We’re a little obsessed with cells,” said Alfa when speaking at NVIDIA GTC in San Jose last week. “If we’re training advanced AI models, we care about tokens that are produced by the underlying data.”
Noetik takes a human-first approach to predict cancer clinical outcomes using multimodal datasets from patient-derived tumors. This strategy has attracted GSK, which earlier this year entered a five-year licensing partnership to access Noetik’s non–small cell lung cancer and colorectal cancer foundation models.
Tessier-Lavigne sees X-Cell as just the beginning of a long term program that will be built “brick by brick.”
“For any company or lab, you balance creating the ultimate model with being pragmatic about gaining insights in the near term.” he said. Notably, investing in large-scale Perturb-seq datasets enables learning underlying gene regulatory networks that drive biological function across the genome.
Wang says Xaira has started with the virtual cell but looks toward modeling more complex systems, including animals, organoids, and eventually patients. As generating patient level data remains time-consuming and expensive, scalable cellular models can act as a bridge that generates hypotheses to be tested in patient AI models.
The Xaira team also plans to expand the work to more modalities, such as chemical perturbations that modulate signaling pathways and proteomic data. In this vein, Biohub, Arc Institute, and Tahoe Therapeutics announced a new team effort in January to build a massive chemical perturbation dataset to be released open source. The timeline for data release has not been disclosed.
“The beauty of AI is creating a foundation where you can add more data and have transfer learning across different dimensions,” says Tessier-Lavigne. “That is our aspiration.”
As single-cell data capture only one slice of biological complexity, diverse data modalities are not competing, but complementary. Each advance adds a new layer of resolution, steadily moving the field toward a more complete and predictive model of biology.
The post Xaira’s First Virtual Cell Model Is Largest To-Date, Toward Complex Biology appeared first on GEN – Genetic Engineering and Biotechnology News.
GEN – Genetic Engineering & Biotechnology News
Bioengineered Implants Deliver Multi-Drug Therapy in Animal Models
In a new paper, scientists from Northwestern University and their collaborators at Rice University and Carnegie Mellon University report on their progress towards developing so-called implantable “living pharmacies.” These are tiny devices containing engineered cells that continuously produce medicines inside the body. Details of the study, which was done in rats, are published in Device in a paper titled “Design of a wireless, fully implantable platform for in-situ oxygenation of encapsulated cell therapies.”
The device, which is called the hybrid oxygenation bioelectronics system for implanted therapy or HOBIT, is roughly the size of a folded stick of gum. It integrates engineered cells with oxygen-producing bioelectronics and is designed in such a way that the cells are shielded from the body’s immune system while also receiving oxygen and nutrients needed to keep them alive and producing drugs for several weeks. In the future, these devices could be deployed to treat chronic conditions without requiring patients to carry, inject, or remember to take medications.
“This work highlights the broad potential of a fully integrated biohybrid platform for treating disease,” said Jonathan Rivnay, PhD, a professor of biomedical engineering and materials science and engineering at Northwestern and a co-principal investigator of the project. “Traditional biologic drugs often have very different half-lives, so maintaining stable levels of multiple therapies can be challenging. Because our implanted ‘cell factories’ continuously produce these biologics, keeping the cells alive with our oxygenation technology allows us to sustain steady levels [of] multiple different therapeutics at once.”
Solving the oxygenation challenge was critical to the success of HOBIT. When engineered cells are packed together in an implant, they compete for oxygen to live. Without a steady supply, many cells die, which limits how much medicine the implants can produce. In an earlier study, Rivnay and his collaborators demonstrated how a tiny electrochemical device could generate oxygen by splitting nearby water molecules, and showed that supplying oxygen locally dramatically improved the survival of implanted therapeutic cells. The latest iteration of their device integrates that oxygen-generation technology in a fully implantable, wireless system.
Digging into the details of the device, HOBIT contains three primary components: a cell chamber that holds the genetically engineered cells, a miniature oxygen generator, and electronics and a battery to regulate oxygen production and wirelessly communicate with external devices. Because the device produces oxygen directly inside the implant, the cells receive a steady supply even in hypoxic environments. “We are producing oxygen directly where the cells need it,” Rivnay said. “That allows us to support much higher cell densities in a much smaller space.” In fact, “cell densities in HOBIT were roughly six times higher than conventional unoxygenated encapsulation approaches.”
According to the paper, the team engineered the cells to produce three different biologics—an anti-HIV antibody, a GLP-1-like peptide used to treat type 2 diabetes, and leptin, a hormone that regulates appetite and metabolism. They implanted the devices under the skin of rats and monitored drug levels in their bloodstreams for 30 days. Blood measurements of animals with the implanted devices showed sustained levels of all three biologics throughout the study period. In contrast, in animals that were implanted with devices without oxygenation, the biologics that had shorter half-lives were undetectable by the seventh day. Drugs with longer half-lives in these animals also declined steadily over time. At the end of the testing period, roughly 65% of the cells in the oxygenated devices remained viable compared with roughly 20% in control devices.
For their next steps, the scientists intend to test their devices in larger animal models and explore disease-specific applications, including therapies based on transplanted pancreatic cells. “As these technologies continue to develop, devices like this could eventually act as programmable drug factories inside the body—delivering complex therapies in ways that simply aren’t possible today,” Rivnay said.
The post Bioengineered Implants Deliver Multi-Drug Therapy in Animal Models appeared first on GEN – Genetic Engineering and Biotechnology News.
GEN – Genetic Engineering & Biotechnology News
Gut-Immune Link Identified in Multiple Sclerosis-Related Neuroinflammation
Multiple sclerosis (MS) is a debilitating neurological disorder caused by malfunctioning immune responses that target the brain and spinal cord of the central nervous system (CNS). New research led by Shohei Suzuki, MD, PhD, assistant professor, division of gastroenterology and hepatology, and Tomohisa Sujino, PhD, associate professor, School of Medicine, at Keio University, Japan, has now indicated how the gut can initiate neuroinflammation in multiple sclerosis.
Their study found that intestinal epithelial cells (IECs) promote the development of pathogenic T cells that migrated to the spinal cord and induced disease symptoms in mouse models of the disorder.
The researchers examined intestinal tissues from patients with MS and mice with experimental autoimmune encephalomyelitis (EAE), a close analog of MS. In both cases, they observed an increase in TH17 cells and an upregulation of major histocompatibility complex class II (MHC II) expression in IECs. Deleting MHC II in IECs reduced the accumulation of TH17 cells in the gut and lowered the severity of EAE. They suggest the results could inform future strategies for developing targeted therapeutics against autoimmunity.
“While current therapies for MS often target B cells, our study highlights the gut as an important therapeutic site,” Suzuki commented. “Modulating intestinal microbiota or antigen-presenting activity of IECs represents new approaches to treating autoimmune neurological diseases.”
Suzuki, Sujino, and colleagues reported on their findings in Science Immunology, in a paper titled “Intestinal Epithelial MHC Class II Induces Encephalitogenic CD4⁺ T Cells and Initiates Central Nerves System Autoimmunity,” in which they concluded, “Our findings reveal an interaction between gut IECs and neuroinflammatory diseases through MHC II expression in human MS and mouse EAE, providing a mechanistic link between gut immune education and CNS autoimmunity and opening new avenues for targeting intestinal immunity in neuroinflammatory diseases.
Failure of the immune system to distinguish ‘self’ from ‘non-self’ entities leads to excessive autoimmune responses against self-proteins like myelin, which forms a protective covering on the neurons. Multiple factors influence the onset and progression of MS, including genetic susceptibility, environmental triggers, and, more recently, the gut microenvironment. Patients with MS exhibit alterations in their gut microbiota, while the gut microbiota and microbial metabolites play a pivotal role in shaping the chronic autoreactive immune responses. “… in an experimental autoimmune encephalomyelitis (EAE) model, commensal or specific microbes were found to be essential for disease initiation and progression,” the authors wrote.
However, in trying to define this gut–CNS axis, the cellular mechanisms that relay the gut-derived signals to the immune system to influence autoimmune inflammation in the CNS remain poorly understood. “Increasing evidence shows that the gut microbiota influences neurological diseases such as Parkinson’s, Alzheimer’s, and MS,” Sujino stated. “However, the mechanisms linking gut microbes, intestinal immunity, and brain inflammation remain unclear. We were keen to identify how gut immune responses contribute to neuroinflammatory diseases.”
Prior research has shown that gut-derived signals can promote the differentiation of T cells into pathogenic T helper 17 (TH17) in mouse models of MS. Recent studies have suggested that IECs can function as antigen presenting cells that help induce these pathogenic cells, but the underlying mechanisms have been unclear.
Building on their previous observation that mild intestinal (ileal) inflammation exists in experimental autoimmune encephalomyelitis (EAE), which is a mouse model of MS, the authors set out to test whether similar inflammation is present in patients with MS. By performing single-cell RNA sequencing on intestinal biopsies, the team identified that inflammatory Th17 cells accumulate in the mouse EAE model as well as in the intestine of patients with MS, suggesting a conserved gut–CNS axis that may be active in human diseases.
In both EAE mice and patients with MS, intestinal epithelial cells upregulated antigen presentation pathways. Particularly, epithelial cells in the ileum had higher expression of major histocompatibility complex class II (MHC II) that presents antigens to CD4+ T cells. “Clinically, patients with MS exhibited an increased expression of epithelial MHC II–associated genes and an accumulation of CD4 T cells in the small intestine, suggesting the conservation of this gut-CNS axis in human diseases,” the scientists stated. Experiments showed that selective deletion of MHC II in IECs reduced pathogenic Th17 cell generation and disease severity. “Conditional deletion of MHC II in IECs showed that epithelial antigen presentation was indispensable for the local expansion of pathogenic Th17 cells in the gut and their subsequent migration to the CNS,” the team stated.
![Immunofluorescence analysis was performed on terminal ileum samples from Cnt, IECΔMHCII, Cnt + EAE, and IECΔMHCII + EAE mice. A total of 3–5 tissue sections were analyzed per mouse, with 3 mice included in each group. [Shohei Suzuki]](https://www.genengnews.com/wp-content/uploads/2026/03/low-res-2-1-300x96.jpeg)
mouse, with three mice included in each group. [Shohei Suzuki]
IECs do not typically present antigens to immune cells. So, the team conducted co-culture assays to test the antigen presentation function of IECs. Their findings demonstrate that IECs can directly present antigens in an MHC II-dependent manner to prime CD4+ T cells in the gut. Notably, in these assays, IECs induced Th17 polarization of activated CD4+ T cells. It became clear that the gut was a critical site for immune activation of pathogenic CD4+ T cells that polarized into pro-inflammatory Th17 cells. “These findings provide direct functional evidence that IEC-expressed MHC II is sufficient to drive Th17 polarization from primed CD4 T cells in an antigen-dependent manner, supporting a direct role for IECs as non-professional antigen-presenting cells,” the scientists reported.
To investigate whether the Th17 cells directly contribute to the pool of autoreactive cells in the CNS, they used transgenic mice that express the Kaede protein, which undergoes photoconversion from green to red fluorescence upon exposure to violet light. This model allowed for precise tracking of pathogenic Th17 cells induced in the intestinal lamina propria that then migrate to the spinal cord and drive neuroinflammation.
Taken together, the study findings reveal a critical role for MHC II expressed by IECs in the expansion of pathogenic Th17 cells that subsequently migrate to the CNS during EAE, providing a mechanistic link between gut immune responses and autoimmune neuroinflammatory diseases. The results demonstrate that while systemic circulation allows T cell exchange across immune tissues, the epithelial–immune interactions within the gut mucosal compartment can essentially shape effector T cell responses in the brain.
“This study reveals a previously unknown role of IECs in antigen presentation and Th17 programming, thereby defining a gut-CNS immunological axis with important implications for understanding and treating autoimmune neuroinflammation,” the authors concluded. “Our findings suggest that the modulation of epithelial antigen presentation could serve as a novel therapeutic approach for MS and related diseases. Given the accessibility of the gut epithelium to dietary, microbial, and pharmacological interventions, targeting IEC–T cell interactions may offer a tractable strategy for immunomodulation.”
The post Gut-Immune Link Identified in Multiple Sclerosis-Related Neuroinflammation appeared first on GEN – Genetic Engineering and Biotechnology News.
GEN – Genetic Engineering & Biotechnology News
Agentic AI, Virtual Cell, LNP Vaccine Boosters, Engineered Organs, and Mergers
This week, agentic AI steps into the limelight buoyed by the momentum from generative AI. And there’s a new virtual cell model in town courtesy of AI-drug developer Xaira Therapeutics. From the frontiers of AI, our discussion turned to feats of engineering in regenerative medicine and lipid nanoparticles. In one study, scientists redesigned LNPs to avoid the liver and accumulate in the lymph nodes. In the other, efforts to develop and implant a lab grown esophagus from donor pigs bear fruit. Finally, Novartis plans to spend up to $3 billion to expand its cancer pipeline with the acquisition of Pikavation Therapeutics. And Merck is acquiring Terns Pharmaceuticals for approximately $6.7 billion also with an eye towards boosting its cancer portfolio.
Listed below are links to the GEN stories referenced in this episode of Touching Base:
NVIDIA GTC 2026: Agentic AI Inflection Hits Healthcare and Life Sciences
By Fay Lin, PhD, GEN Edge, March 18, 2026
Xaira’s First Virtual Cell Model Is Largest To-Date, Toward Complex Biology
By Fay Lin, PhD, GEN Edge, March 25, 2026
Modified Lipid Nanoparticles Boost mRNA Vaccine Delivery to Lymph Nodes
GEN, March 24, 2026
Engineered Esophagus Rebuilds Missing Organ Segment in Pig Models
GEN, March 20, 2026
Novartis Acquires Pikavation for Up to $3B, Expanding Cancer Pipeline
GEN, March 22, 2026
Merck Bolsters Cancer Pipeline with $6.7B Terns Buyout
By Alex Philippidis, GEN Edge, March 25, 2026
Touching Base Podcast
Hosted by Corinna Singleman, PhD
Behind the Breakthroughs
Hosted by Jonathan D. Grinstein, PhD
The post Agentic AI, Virtual Cell, LNP Vaccine Boosters, Engineered Organs, and Mergers appeared first on GEN – Genetic Engineering and Biotechnology News.
-
Uncategorized9 years agoThese ’90s fashion trends are making a comeback in 2017
-
Contributors9 years agoThe final 6 ‘Game of Thrones’ episodes might feel like a full season
-
Uncategorized9 years agoAccording to Dior Couture, this taboo fashion accessory is back
-
Uncategorized9 years agoThe old and New Edition cast comes together to perform
-
Uncategorized9 years agoPhillies’ Aaron Altherr makes mind-boggling barehanded play
-
Uncategorized9 years agoUber and Lyft are finally available in all of New York State
-
Uncategorized9 years agoDisney’s live-action Aladdin finally finds its stars
-
Uncategorized9 years agoSteph Curry finally got the contract he deserves from the Warriors