AI x BioPharma Pavilion

Drug discovery often measures biology at the cell level, while therapies must ultimately work across tissues, organs, and whole patients. This scale mismatch means that even highly accurate cellular predictions can fail to translate in the clinic. This session explores strategies to bridge that gap. How do we connect single-cell dynamics to organ-level physiology and patient outcomes? How do we preserve biological context while scaling models? And how do we ensure that virtual biology does not stop at simulation, but informs real therapeutic decisions? Speakers will discuss multiscale modeling that links molecular and cellular systems to higher-order biology; spatial and high-dimensional phenotypic data that retain context; and integrated computational–experimental loops that translate cellular signals into clinically meaningful biomarkers. Together, we ask: how do we ensure virtual biology reflects not just what cells do in isolation, but how biology behaves in the full complexity of patients?

For the first time, AI is enabling us to imagine medicines that “think” - turning on only inside diseased cells or under specific physiological conditions. Neural networks trained on RNA, protein, and cellular data are unlocking a new generation of programmable therapies with unprecedented precision, from cancer drugs that remain inert until encountering tumor signals to RNA medicines capable of adapting to dynamic biological environments. But designing intelligent molecules is only part of the challenge. As AI expands the space of possible therapeutics, the field must also confront a critical question: how do we reliably build, test, and manufacture increasingly complex biological designs? This session explores the emerging continuum from AI-designed molecules to manufacturable programmable therapeutics, examining how advances in sequence design, synthesis, delivery, and validation are translating computational insight into real-world medicines. The future of medicine isn’t static molecules - it’s intelligent, adaptive therapeutics engineered across the full stack, from algorithm to clinic.

The next frontier of biology isn’t in predicting a single static protein structure, but in capturing how proteins move, fold, and interact across time and environments. This session explores how AI can illuminate protein conformations and dynamics, and extend those insights into virtual multi-cellular or tissue models. Experts will discuss the challenge of integrating heterogeneous datasets and instruments, and how breakthroughs in dynamic modeling could reshape drug design, disease understanding, and biomanufacturing. Can we build models that reflect the living, breathing complexity of biology—not just snapshots, but motion?

Why do so many in silico models fail when moved to the lab or clinic? Too often, they’re trained on incomplete, non-human, or non-representative datasets. This session tackles the “data gap” head-on: from interoperability bottlenecks and the black box problem to the limits of current virtual cell simulations (~50 million perturbations vs. the billions biology demands). Panelists will explore how to create “human-first” datasets that reflect real biology, unlock mechanistic interoperability, and close the discovery–development divide. The goal: build AI tools that can directly identify viable drug candidates instead of stalling in silico.

Biology is entering an era where AI agents don’t just analyze data — they co-design, plan, and execute experiments. Multi-agent systems like CRISPR-GPT demonstrate how AI can act as a true lab co-pilot: decomposing complex genome editing projects into stepwise workflows, selecting tools, troubleshooting, and even drafting protocols that allow junior researchers to perform sophisticated edits on their first attempt . Beyond CRISPR, new systems like BioMARS integrate reasoning agents with robotics, while biotech companies are testing “AI lab assistants” that monitor and adjust experiments in real time. This session explores how multi-agent copilots are making biology more reproducible, democratizing complex workflows, and pushing the boundaries of lab autonomy. The central question: when AI can plan, troubleshoot, and validate experiments end-to-end, how should scientists and institutions govern this new power?

Biology is entering an AI-driven era, but most experimental infrastructure still produces data designed for individual experiments, not for learning at scale. As a result, much of today’s data is useful in the moment but poorly suited for training robust, long-lived models. This session will explore what biological data matters most today, what data needs to be generated now to support future models, and how leading teams are closing that gap. Panelists will discuss how automation, metadata discipline, and standardized testing pipelines can turn artisanal lab workflows into continuous experiment-to-learning systems. The focus will be on infrastructure and experimental design, highlighting practical bottlenecks, emerging best practices, and what becomes possible when biology produces abundant, high-quality, model-ready data by default.

From tissues morphing in development to microbes competing in a bioreactor, biology is inherently emergent. Multi-agent simulations — from platforms like BioDynaMo, CompuCell3D, and BIO-LGCA — are now powerful enough to model billions of interacting agents, capturing diffusion, metabolism, migration, and signaling with physical fidelity. Synthetic biologists are using these frameworks to probe design limits before moving to the lab, asking questions like: How far can diffusion alone carry a signaling molecule? What metabolic bottlenecks emerge in crowded cells? And how do engineered traits play out at population scale? This session will spotlight how agent-based models are becoming essential design environments for synthetic biology, helping teams test hypotheses virtually, anticipate failure modes, and translate biology into an engineering discipline rooted in predictive, quantitative simulation.

AI is reshaping how biopharma discovers, develops, and advances therapeutic agents across the full lifecycle, from early design to translational strategy and clinical asset development. But with dozens of platforms and models emerging, R&D leaders face a strategic crossroads: should they build internal AI capabilities, buy turnkey software, or partner with integrated platforms that connect computational design, experimental validation, and clinical decision-making? This session brings together Biotech R&D executives and AI platform leaders to explore how software-first, closed-loop AI workflows are transforming not only discovery speed, but also translational success and clinical outcomes. Speakers will share real-world perspectives on integrating AI into portfolio strategy, advancing assets toward the clinic, repositioning clinically validated assets, and redefining the operating model for biologics development.

As biology becomes programmable and AI accelerates discovery, startups are generating breakthrough innovations at unprecedented speed. Yet translating these advances into real-world therapies still depends on effective collaboration with global pharmaceutical organizations. This session explores how the innovation ecosystem connects early-stage breakthroughs to scalable development, bringing together leaders from startup incubation, external innovation, and pharma strategy. Speakers will examine how AI-native biotech companies engage with pharma today: how startups become “pharma-ready,” how external innovation teams evaluate and structure partnerships, and what collaboration models are emerging as biology and computation converge. From early ecosystem support and venture building to strategic alliances and co-development pathways, the discussion will provide a practical look at how ideas move from discovery to patient impact in the AI era.

This talk will introduce the development of artificial Agents to model biological phenomena in molecular biology, biotechnology, and synthetic biology incorporating reinforcement learning, differential equation modeling of molecular dynamics, and agentic bio-causal reasoning. Agent to agent interaction with the A2A and PoR protocols, and MCP and API interfaces to Machine Learning (Neural Network) Models including causal reasoning models and bio-specific models will be discussed. Synthetic biology deals with huge possibility spaces in terms of the combinatorics of nuceotide and proteomic sequences in proposed novel genes and proteins and how to constrain possibility spaces into computable functional novel genes, genetic circuits, gene regulatory networks and novel functional proteins will be discussed. Hence the sheer complexity of biological phenomena requires advanced Agentic AI and machine learning models to efficiently process, find patterns in, and reason about these complex systems with hundreds of thousands of variables, millions of connections, and potentially trillions of parameters. The current state of Agentic Bio research will be covered and where the research needs to go will be elucidated. Finally an application of Agentic Inter and Intra-cellular Signaling will be presented in detail to see the nuts and bolts of how Agentic AI can model a biological phenomenon with molecular biological, medical, and synthetic biological applications. The presenter’s background includes advanced degrees in computer science and computational molecular biology with experience in bio-computational modeling including a computational neuroscience project at Stanford where the neurogenetic and synaptic development of the C.elegans’ brain was modeled. Synthetic Biology: the possibility spaces are endless!

Biology will be the center of the next industrial revolution, representing a $4 trillion economic opportunity. Yet, this value remains overwhelmingly unrealised for one fundamental reason: nature never intended to power industrial manufacturing. Biology was optimized for survival, not for the high-efficiency processes required to transform the global economy. For too long, the industry has relied on incremental improvements, essentially duct-taping enzymes and calling them industrial. At Biomatter, we believe that complete freedom to design any enzyme is the only way to realize the full potential of biomanufacturing. By combining Generative AI with rigorous physics engines, our Intelligent Architecture™ platform allows us to step outside the bounds of natural selection and build enzymes from the bottom up. We are turning the "previously impossible" into routine. From liberating enzymes of their cofactor dependencies for mRNA raw materials to designing lactases that reject the trade-off between lactose removal and high GOS fiber formation, we are proving that biology’s limits are negotiable. Join us to see how we are building the enzymes nature couldn't.

While Twist Bioscience may look like an overnight success, its rise reflects a decade of persistence, innovation, and platform building. In this main stage keynote, CEO and co-founder Emily Leproust shares the journey from startup vision to global leader in DNA synthesis and programmable biology, highlighting lessons learned scaling deep technology, navigating industry cycles, and building trusted infrastructure for biotech and pharma. Looking ahead, Twist is positioning itself at the forefront of the convergence between AI and biology, using DNA as an information layer to accelerate drug discovery and advance human health. This keynote explores how long-term thinking and bold ambition are shaping the next era of AI-driven therapeutics.

Current bioprocess monitoring is limited to basic environmental proxies like pH and dissolved oxygen. Schmidt Sciences is changing this paradigm by adapting advanced physics for biology. This talk introduces three cutting-edge sensing platforms currently in development: fluorescent nanodiamonds, single-cell Raman spectroscopy, and non-invasive optical frequency combs. Join us to learn how these high-dimensional data streams are being integrated with machine learning to predict campaign outcomes and revolutionize how we monitor cell health at scale.

The PURE system, invented 25 years ago, established a fully reconstituted approach to cell-free protein synthesis. What began as a system to better understand translation has evolved into a versatile platform for engineering biology. This talk highlights how PURE-derived platforms such as PUREfrex® enable rapid prototyping, high-throughput screening, and AI/ML-driven optimization, accelerating synthetic biology and next-generation biologics development.

AI-enabled, self-driving labs are still emerging, but their foundations are already transforming how teams design, run, and interpret experiments. This session offers a practical guide for scientists and R&D leaders who want to understand what can be done today — from tightening design–test–learn loops to reducing manual error and capturing early benefits of autonomous experimentation. Rather than presenting an unrealized future, speakers will focus on practical, real-world steps that give organizations a competitive edge as SDL capabilities evolve and mature. Speakers will explore what’s working, what’s not, and how autonomous lab systems are reshaping protein engineering, pathway optimization, and therapeutic design.

Biology has long relied on a limited molecular vocabulary shaped by natural evolution. Today, that alphabet is expanding. Advances in expanded genetic codes, non-canonical amino acids, macrocycles, de novo design, and AI-guided protein engineering are enabling scientists to create biomolecules with properties and functions that nature never evolved. This session explores the rise of programmable biomolecules at the intersection of biology, chemistry, and computation. Rather than simply optimizing existing proteins, researchers are building entirely new classes of functional molecules with novel architectures, chemistries, and therapeutic potential. From next-generation biologics to hybrid molecular scaffolds, the discussion will examine how the field is moving beyond nature’s defaults and toward a future where biomolecules can be designed with increasing precision, flexibility, and intent.

Standard bioreactors often lack the instrumentation required to rapidly monitor cell physiology, leaving critical gaps in our understanding of scale-up dynamics. This session presents active projects from the Schmidt Sciences’ Sensors for Biomanufacturing Program designed to address this challenge through novel sensing modalities. Spanning from near real-time intracellular measurements to non-invasive off-gas fingerprinting, the panel brings together technology developers and industrial bioprocess experts to discuss the translation of these tools from the lab to the plant floor. Together, we will critically evaluate the utility of high-dimensional metabolic data and explore the engineering requirements for integrating physics-based sensors and machine learning into existing biomanufacturing workflows.

Drug discovery is not limited by models. It is limited by data. While AI is accelerating molecular design and target discovery, the real bottleneck remains the generation, integration, and interpretation of biological datasets that are complex, heterogeneous, and often not yet predictive. Pharma-scale discovery requires more than algorithms. It requires new approaches to building and operationalizing data itself. This session explores how next-generation DNA synthesis, high-throughput experimentation, and integrated data infrastructures are enabling a new biology data flywheel. From experimental datasets that inform translational decisions to emerging standards for capturing real-world and preclinical signals, leaders will discuss how data generation strategies are reshaping discovery workflows. Speakers from pharma, AI-native biotech, and platform providers will examine how biology is becoming a programmable data layer, enabling faster biologics development, more informed portfolio decisions, and new collaborative models that connect experimental systems, computational tools, and pharma-scale discovery pipelines.

The Design–Build–Test–Learn (DBTL) cycle remains the core engine of biological engineering, yet its iteration speed still lags far behind software development. As AI systems begin to design, plan, and execute experiments, a new paradigm is emerging: DBTL as an autonomous, continuously optimizing system. Next-generation platforms combine AI-assisted rational design, high-throughput construction and perturbation, real-time data acquisition, and active learning to close the loop at unprecedented scale. Agent-powered lab-in-a-loop workflows, lab-on-a-chip systems, and advances at the silicon-to-carbon interface are enabling tighter integration between computation and biology, from semiconductor-enabled sensing to real-time feedback and decision-making. This session explores how autonomous DBTL stacks could unlock software-like iteration velocity in biology, redefine experimentation, and reshape the future of programmable discovery.

Brain Computer Interfaces (BCIs) hold immense promise to help restore critical functions now for individuals with neurological conditions, severe speech impairments, and paralysis. Over the last thirty-five years, major advancements in artificial intelligence, brain mapping, and material sciences are laying the foundation for a future where BCI-enabled augmented experience is as common as accessing the internet or using a mobile phone. Join Paradromics CEO Matt Angle, PhD to discuss the latest on neurotechnology today, as well as expansive future BCI applications.

For more than a century, everyday products - from detergents and shampoos to textiles and packaging - have relied on petrochemicals and harsh industrial processes. Today, AI-driven protein design is opening a radically different path: creating custom enzymes and biomolecules that outperform traditional chemistry while reducing environmental impact. This session explores how advances in computational protein design and machine learning enable the rational creation of enzymes tailored for home care, personal care, and next-generation materials—moving beyond incremental discovery to purpose-built performance under real industrial conditions. Critically, this highlights how AI-driven design is being translated into commercially deployed products at scale with partners and customers.

Biologics and engineered proteins have traditionally evolved through cycles of intuition, screening, and incremental optimization. Today, AI is transforming proteins into programmable systems; governed by learnable patterns across activity, stability, expression, specificity, manufacturability, and environmental performance. This shift is unlocking a new generation of biomolecules, from next-generation therapeutics to sustainable enzymes and functional biological systems, that would have been impossible to design by hand. In this session, leaders from biopharma, industrial biotech, machine learning, and protein engineering will explore how multiparameter optimization, generative modeling, and closed-loop experimental validation are reshaping biomolecular design across domains. From clinical biologics to planetary-scale applications, we examine the shift from trial-and-error to predictive, constraint-driven design, and what it means for R&D timelines, scalability, and real-world impact.