Virtual cells go multiscale to predict complex biology. Posted on July 24, 2026 By News Staff Virtual cell models that enable the prediction of cell behavior across scales and biological contexts are rapidly emerging at the forefront of drug discovery. Tom Sercu, PhD, vice president of AI and engineering at Biohub, points to a clinician-scientist studying a rare autoimmune disease as an example of how AI models could reshape translational medicine. Starting from a patient's genome, researchers could use virtual cells to predict how major immune cell types behave in disease versus healthy states. The result offers an invaluable tool across target and mechanism-of-action discovery, patient stratification, toxicity prediction, and therapeutic development. Yet, building a virtual cell is not an easy feat. "Transformative AI in biology does not come from algorithms alone, but when models are trained on large-scale, high-quality, openly accessible datasets," says Sercu. To capture complex biology, such data must span model systems and organisms, interventional and observational methods, and diverse cellular states. To support this mission, Biohub announced a $500 million commitment to the Virtual Biology Initiative in April. The five-year campaign will accelerate the generation of technologies and multi-modal datasets needed to power virtual cell models. Similar to how more than 253,000 experimentally determined molecular structures in the Protein Data Bank (PDB), assembled over five decades, became foundational training data for modern AI protein-structure prediction, Sercu sees an analogous moment for cellular biology. "We do not yet have the equivalent of the PDB for cells," he emphasized. "The Virtual Biology Initiative seeks to change that." Today's virtual cell developers reflect on what's needed for these models to predict complex biology and overhaul drug discovery. Single or bulk. Much of the industry has defined the virtual cell as transcriptome models that predict how perturbations alter gene expression across cellular contexts. Among the increasingly crowded ecosystem, Arc Institute's first-generation virtual cell model, STATE, predicts how stem cells, cancer cells, and immune cells respond to drugs, cytokines, or genetic perturbations. In March, billion-dollar-backed, Xaira Therapeutics unveiled X-Cell, the first scaling law demonstrator in the virtual cell domain, sizing up to a whopping 4.9 billion parameters. These models aim to generalize to unseen biological contexts by training on causal single-cell RNA sequencing (scRNA-seq) data. 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. Ginkgo Datapoints, the AI platform division of Ginkgo Bioworks, looks toward bulk transcriptomics rather than a single-cell approach. "Just like how models benefit from diversity in training data, we as an industry benefit from having diversity of approaches," said John Androsavich, PhD, general manager at Ginkgo Datapoints. The Datapoints team applies high-throughput automation to create diverse biological datasets, including cell perturbations, antibody developability, and ADME small molecule developability data, to support AI model training for life science partners. In March, Ginkgo Datapoints delivered the first data release of the Virtual Cell Pharmacology Initiative (VCPI). Approximately 2,280 small molecules were profiled in full dose response using DRUG-seq, a scalable arrayed transcriptomics assay measuring chemical perturbations. In contrast to scRNA-seq, which covers approximately 1,500 genes per cell, DRUG-seq captures nearly 10,000 genes per condition with higher signal-to-noise to optimize insights for pharmacology. Notably, VCPI has exclusively focused on THP-1, a human monocytic cell line widely used across immunology, oncology, and inflammatory disease research, to understand drug action. Across space. While the crowding around scRNA-seq has largely been driven by the pursuit of scale, "a cell is not only its RNA," tempers Hani Goodarzi, PhD, core investigator at Arc Institute. He emphasizes that cells are complex systems shaped by multiple layers of biology beyond gene expression alone, including protein abundance, chromatin state, spatial organization, metabolism, and post-translational regulation. A useful analogy comes from large language models (LLMs), which became powerful as text provided an exceptionally scalable substrate for training trillions of tokens. Yet, text alone is an incomplete representation of human communication. "The lesson is not that one modality is sufficient forever," says Goodarzi, "but that a single high-quality, scalable modality can support general representations when the training corpus is large." Emma Lundberg, PhD, co-founder and CSO at GenBio AI, is worried about the "streetlight effect." "We're scaling what we can and not necessarily what we should," she says. GenBio AI seeks to develop world models that cross multiscale biology. Instead of concentrating on one data modality, the company's so-called "AI-Driven Digital Organism" grows expertise in embedding, tokenizing, and training models across scales, from the molecular layer to regulatory networks. Rather than undergo internal data generation, GenBio AI focuses on public data and partnerships to power the company's models. Lundberg, who is also associate professor of bioengineering and pathology at Stanford University, argues that models can guide the field to which data modalities to pursue. As an example, models that incorporate biological priors, such as protein-protein interactions, can achieve noticeable improvements in predictive performance. Spatial and temporal data also capture critical dimensions of biological function that sequence data alone cannot resolve. According to the Human Protein Atlas, roughly 60% of human genes encode proteins that localize to multiple cellular compartments, often carrying out distinct functions depending on context. In a May preprint posted on bioRxiv, Lundberg and colleagues introduced ProtiCelli, a deep generative model that visualizes the spatial organization of nearly the entire proteome within individual cells. By training on 1.23 million images from the Human Protein Atlas, the model simulates microscopy images for 12,800 human proteins while also generalizing to unseen cell types and drug perturbations absent from training. Through time. Cellular Intelligence is developing a universal virtual cell signaling model designed to simulate cell-state transitions over time, with the goal of expanding the possibilities of regenerative medicine. By learning the underlying "grammar" through which sequences of signaling cues drive cell differentiation, these models aspire to enable the on-demand generation of any cell type. Less than one percent of known human cell types can be reliably produced for downstream applications in cell therapy. As only 20 fundamental molecular signaling pathways give rise to thousands of cell states, researchers face an unfathomably large search space when engineering a particular cell type. "Every cell that we discover or optimize opens a slew of potential applications," said Micha Breakstone, CEO and co-founder of Cellular Intelligence. "One could spend a decade and tens of millions of dollars on painstaking trial-and-error to differentiate a new cell type, or solve this problem in one fell swoop, much like AlphaFold for the protein folding challenge." The company's platform leverages a semi-permeable capsule technology, which selectively retains cells and large analytes while being freely accessible to media, enzymes, and reagents. The method enables high-throughput assays combining live-cell culture with genome-wide readouts. Millions of time-varying signal combinations are tested on human stem cell differentiation in parallel, providing 1,000 times higher efficiency than traditional methods. In May, Cellular Intelligence advanced as a Phase II-ready clinical company after entering an agreement with Novo Nordisk to acquire STEM-PD, an allogeneic cell therapy program for Parkinson's disease with Fast Track Designation. The deal comes six months after Novo announced its strategic exit from the cell therapy space. The start-up's AI cell signaling models will address protocol development, one of the biggest obstacles preventing cell therapies from clinical impact. "Novo selected Cellular Intelligence as the right partner because the next major challenge for complex cell therapy programs is not only the biology," says Breakstone. "It is manufacturing scale-up, comparability, clinical logistics, and commercial readiness." As the diversity of virtual cell models targets new dimensions of complex biology, every approach takes another step closer toward clinical impact.

GenBio AI advances agentic AI for building virtual-cells - powered by NVIDIA. June 23, 2026 - Today, GenBio AI announced its collaboration with NVIDIA to accelerate the development of virtual-cell world models: AI systems designed to simulate human cellular behavior across biological modalities and scales. What is a virtual-cell world model? To discover and develop new medical treatments, researchers must understand how living human cells respond to a wide variety of changes: what happens throughout the system when a gene is switched off, a drug is introduced, or a disease disrupts normal function? With conventional methods, this is extraordinarily difficult. Of every 10,000 compounds entering the drug development pipeline, only one on average reaches the clinic, because human biology is incredibly complex and existing tools capture only fragments of it. A virtual cell world model (VCWM) is a simulation engine built to solve this problem. Using world model architecture, biologists will be able to predict, simulate, and ultimately program cellular processes across all biological scales, molecular and cellular. This enables deeper understanding of disease mechanisms, faster testing of therapeutic hypotheses, and better evaluation of candidate interventions in a digital laboratory before any physical experiment is run. GenBio AI's central thesis is that the virtual cell is not simply another predictive model. A true virtual cell is a computational system that can reason over biological state, intervention and outcome, shifting biology from isolated prediction tasks toward comprehensive simulation. The gap between that vision and today's reality is stark. Building one good cell model requires researchers to spend countless hours manually designing architectures, tuning dozens of settings, fixing problems, and testing ideas. This process can take months of expert iteration with no guarantee of a useful result. To close that gap, the GenBio AI team built a new tool: VCHarness. What is GenBio AI's VCHarness? VCHarness is GenBio AI's autonomous agentic system for constructing virtual-cells. Rather than relying on human researchers to design each model by hand, VCHarness does the work autonomously: it proposes a candidate model, writes the code, runs the experiment, measures the results, learns from what worked and what didn't, and tries again - all in a continuous loop. The system combines a library of pre-trained biological foundation models (covering the DNA, protein, RNA, and cellular levels) with AI coding agents, structured search, evaluation, memory, and distributed execution to generate, test, debug, and refine candidate virtual cell components. Together, these allow VCHarness to efficiently search over complete executable modeling workflows and learn from experimental results, going from biological question to validated model candidate in a fraction of the time previously required. In tests predicting how human cells respond to CRISPR gene edits (a benchmark task for virtual-cell research), VCHarness discovered models that outperformed expert-designed baselines in days rather than months. Critically, it didn't just tune existing designs - it discovered novel combinations of biological AI components. How will GenBio AI and NVIDIA work together? GenBio AI is integrating NVIDIA BioNeMo, NVIDIA Megatron, NVIDIA NIM microservices, and the NVIDIA BioNeMo Agent Toolkit to further accelerate the construction and deployment of virtual-cell systems. GenBio AI plans to leverage NVIDIA's life-sciences AI infrastructure, model-development tooling, and agent-ready biology components as part of a broader stack for scaling autonomous model construction, and high-throughput virtual cell experimentation. "GenBio AI is building toward a functional world model of the virtual cell: an AI system that can simulate how cells respond to a wide variety of genetic, chemical, and environmental interventions," said GenBio AI President and Chief Scientist Eric Xing. "VCHarness is our agentic engine for building these systems. NVIDIA BioNeMo and the BioNeMo Agent Toolkit provide crucial infrastructure and capabilities that allow us to scale this work faster, while our internal teams continue to develop the core models, algorithms, and autonomous systems that define our virtual-cell platform." By combining GenBio AI's virtual-cell world-model architecture and autonomous model-building technology with NVIDIA's accelerated computing and BioNeMo ecosystem, the companies aim to make agentic life-sciences workflows more scalable, reproducible, and useful for real-world discovery. About GenBio AI. GenBio AI is building AI systems for biology, with a focus on world models of the virtual cell. The company develops biological foundation models, autonomous agentic systems, and multimodal simulation frameworks designed to predict, simulate, and program cellular behavior across scales.

PALO ALTO, Calif., June 17, 2025 /PRNewswire/ -- GenBio AI, a company building the World's First AI-Driven Digital Organism (AIDO), is proud to announce that Dr. Emma Lundberg has formally joined the company as Co-Founder and Chief Scientific Advisor.In this role, Lundberg will focus on the development of GenBio AI's data strategy and other scientific initiatives, advancing its mission to build a generative digital organism