Tag:

Healthcare and Life Sciences

Tech

Lilly & NVIDIA Launch AI Factory for Drug Discovery | NVIDIA Blog

by Chief Editor
written by Chief Editor

The AI Revolution in Drug Discovery: Beyond LillyPod

The launch of LillyPod, powered by over 1,000 NVIDIA Blackwell Ultra GPUs, marks a pivotal moment in pharmaceutical innovation. But it’s not simply about computational power; it’s about fundamentally changing how drugs are discovered, developed, and delivered. This new era promises faster timelines, reduced costs, and, more effective treatments.

The Rise of AI Factories in Pharma

LillyPod isn’t an isolated case. Pharmaceutical companies are increasingly investing in dedicated AI infrastructure – “AI factories” – to accelerate research. These facilities leverage the latest advancements in accelerated computing, networking, and AI software to handle the massive datasets and complex models required for modern drug discovery. The goal is to move beyond traditional, trial-and-error methods to a more predictive and efficient approach.

Foundation Models: The New Building Blocks

A key driver of this transformation is the emergence of foundation models. These large AI models, trained on vast amounts of data, can be adapted to a wide range of tasks, including protein structure prediction, small-molecule design, and genomics analysis. Lilly’s use of these models, coupled with NVIDIA FLARE, allows for collaborative research while maintaining data privacy.

Pro Tip: Federated learning, enabled by technologies like NVIDIA FLARE, is crucial for pharmaceutical companies seeking to collaborate on AI development without compromising sensitive patient data.

Accelerating Genomics and Personalized Medicine

The ability to analyze genomes at scale is unlocking new possibilities for personalized medicine. LillyPod’s capacity to process 700 terabytes of data with over 290 terabytes of high-bandwidth GPU memory will enable researchers to identify genetic markers associated with disease and develop targeted therapies. This represents a shift from treating symptoms to addressing the root causes of illness.

The Impact on Clinical Trials

AI is also poised to revolutionize clinical trials. By analyzing patient data and predicting trial outcomes, AI can help optimize trial design, identify suitable candidates, and reduce the time and cost associated with bringing new drugs to market. This could lead to faster access to life-saving treatments for patients in need.

Beyond Discovery: AI in Manufacturing and Supply Chain

The benefits of AI extend beyond the initial stages of drug discovery. AI-powered systems can optimize manufacturing processes, improve quality control, and enhance supply chain efficiency. This ensures that drugs are produced reliably and delivered to patients on time.

The Role of Networking and Infrastructure

The success of AI factories like LillyPod relies on robust networking infrastructure. Technologies like NVIDIA Spectrum-X Ethernet are essential for enabling high-speed data transfer and communication between GPUs, ensuring optimal performance. Efficient liquid cooling is also critical for managing the energy demands of these powerful systems.

Future Trends to Watch

  • Agentic AI: The development of AI agents capable of autonomously designing and executing experiments will further accelerate the discovery process.
  • Generative AI: Generative AI models will play an increasingly important role in creating novel drug candidates with desired properties.
  • Digital Twins: Creating digital twins of patients and biological systems will enable researchers to simulate drug responses and personalize treatment plans.
  • Increased Collaboration: Platforms like Lilly TuneLab will foster greater collaboration between pharmaceutical companies and AI developers.

FAQ

What is an AI factory?

An AI factory is a dedicated infrastructure designed to accelerate AI-driven research and development, typically featuring high-performance computing resources and specialized software.

What are foundation models?

Foundation models are large AI models trained on vast datasets that can be adapted to a variety of downstream tasks.

How does NVIDIA FLARE contribute to AI collaboration?

NVIDIA FLARE enables federated learning, allowing organizations to collaborate on AI projects while keeping their data private.

The launch of LillyPod signals a new era of AI-driven pharmaceutical innovation. As AI technologies continue to advance, we can expect even more breakthroughs in drug discovery and development, ultimately leading to better health outcomes for patients worldwide.

0 comments
0 FacebookTwitterPinterestEmail
Tech

NVIDIA & Dassault Systèmes Partner to Build Industrial AI World Models

by Chief Editor
written by Chief Editor

The Rise of Virtual Twins: How AI is Revolutionizing Engineering and Manufacturing

The future of engineering isn’t about building physical prototypes first – it’s about building them in software. A landmark partnership between NVIDIA and Dassault Systèmes, unveiled at 3DEXPERIENCE World, is accelerating this shift, promising to redefine how products are designed, factories are operated, and even scientific discoveries are made.

From Digital Designs to ‘World Models’

For decades, engineers have used digital models to visualize and test designs. Now, the focus is moving towards “world models” – AI-powered systems that simulate the behavior of products, factories, and complex systems with unprecedented accuracy. These aren’t just static representations; they’re dynamic, physics-based simulations capable of predicting outcomes and optimizing performance.

Dassault Systèmes, with its 3DEXPERIENCE platform serving over 45 million users, has long been a leader in virtual twin technology. The collaboration with NVIDIA aims to fuse accelerated computing and AI libraries with these virtual twins, enabling real-time digital workflows and AI companions to assist engineering teams.

AI as Infrastructure: The New Computing Stack

NVIDIA CEO Jensen Huang envisions a future where artificial intelligence is as fundamental as electricity or the internet. This means moving away from manually specified designs to systems that can generate, simulate, and optimize solutions in software at an industrial scale. This represents a fundamental reinvention of the computing stack.

According to Huang, this new approach will allow engineers to function at a scale 100 to 1,000 times – and eventually a million times – greater than before.

Applications Across Industries

The potential applications of this technology are vast, spanning multiple sectors:

Advancing Scientific Discovery

The NVIDIA BioNeMo platform, combined with BIOVIA science-validated world models, is accelerating the discovery of new molecules, and materials. This has implications for biopharma, materials science, and beyond.

AI-Driven Engineering Design

SIMULIA, leveraging NVIDIA CUDA-X and AI physics libraries, empowers engineers to accurately predict the behavior of designs, enabling faster prototyping and validation. This means fewer physical prototypes and reduced development costs.

The AI-Powered Factory of the Future

NVIDIA Omniverse, integrated with Dassault Systèmes’ DELMIA Virtual Twin, is enabling the creation of autonomous, software-defined production systems. This represents a shift from static factories to dynamic, adaptable manufacturing environments.

Virtual Companions for Engineers

The 3DEXPERIENCE agentic platform, powered by NVIDIA AI technologies and Nemotron open models, will provide engineers with “virtual companions” – AI assistants that offer trusted, actionable intelligence and automate repetitive tasks.

Deploying AI Factories with Sovereign Cloud

Dassault Systèmes is deploying NVIDIA-powered AI factories on three continents through its OUTSCALE sovereign cloud. This allows customers to leverage the power of AI although maintaining data residency and security, addressing critical concerns for many organizations.

Amplifying, Not Replacing, Human Ingenuity

Both Dassault Systèmes CEO Pascal Daloz and NVIDIA CEO Jensen Huang emphasized that the goal isn’t to replace engineers, but to amplify their capabilities. By automating exploratory tasks and providing AI-driven insights, engineers can focus on creativity and innovation.

Daloz stated that engineers want to “invent the future,” not simply automate the past.

FAQ

What is a virtual twin? A virtual twin is a digital replica of a physical asset, process, or system. It allows for simulation, analysis, and optimization without the need for physical prototypes.

What are ‘world models’? World models are AI-powered systems that simulate the behavior of complex systems based on physics and scientific principles.

How will this partnership benefit engineers? The partnership will provide engineers with AI-powered tools and virtual companions that automate tasks, accelerate design cycles, and enable exploration of larger design spaces.

Is AI going to replace engineers? No. The focus is on augmenting human capabilities, not replacing them. AI will handle repetitive tasks, allowing engineers to focus on creativity and innovation.

Where can I learn more about this collaboration? You can explore demos and learn more at GTC San Jose from March 16-19, specifically at Florence Hu-Aubigny’s session on virtual twins and booth 1841 in the Industrial AI and Robotics pavilion.

Did you realize? Virtual twins are becoming “knowledge factories” – places where knowledge is created, tested, and trusted before anything is built in the physical world.

Pro Tip: Explore NVIDIA Omniverse and Dassault Systèmes’ 3DEXPERIENCE platform to understand the capabilities of virtual twin technology and how it can be applied to your industry.

What are your thoughts on the future of AI-powered engineering? Share your insights in the comments below!

0 comments
0 FacebookTwitterPinterestEmail
Tech

NVIDIA & Lilly: $1 Billion AI Lab to Revolutionize Drug Discovery

by Chief Editor
written by Chief Editor

AI Revolutionizes Drug Discovery: A $1 Billion Bet on the Future of Medicine

The pharmaceutical industry is on the cusp of a radical transformation, fueled by the convergence of artificial intelligence and biological research. A groundbreaking collaboration between NVIDIA and Eli Lilly, involving a joint investment of up to $1 billion, signals a new era where drug discovery shifts from painstaking trial-and-error to a data-driven, computationally accelerated process. This isn’t just about faster results; it’s about tackling previously intractable diseases and fundamentally changing how we approach healthcare.

The Power of AI in Modeling Biological Complexity

For decades, drug discovery has been a notoriously slow and expensive process. On average, it takes over 10 years and $2.5 billion to bring a single new drug to market, with a high failure rate. The core challenge lies in the sheer complexity of biological systems. Understanding how molecules interact, predicting drug efficacy, and identifying potential side effects requires navigating an astronomical number of variables.

AI, particularly deep learning, offers a solution. By analyzing vast datasets of genomic information, protein structures, and clinical trial data, AI algorithms can identify patterns and predict outcomes with increasing accuracy. NVIDIA’s expertise in AI computing power, combined with Lilly’s deep pharmaceutical knowledge, aims to create a “scientist-in-the-loop” framework. This means AI won’t replace scientists, but rather augment their abilities, accelerating experimentation and data analysis.

Did you know? AlphaFold, developed by DeepMind, demonstrated the power of AI in protein structure prediction, a critical step in drug discovery. Its ability to accurately predict protein structures has dramatically reduced the time and cost associated with this process.

The Rise of AI Factories and Foundation Models

Lilly’s investment in an NVIDIA DGX SuperPOD – the most powerful AI factory in the biopharma industry – is a testament to this shift. This supercomputer will be used to train large-scale biomedical foundation models. These models, similar to those powering large language models like ChatGPT, will serve as a base for developing specialized AI tools for drug discovery.

These foundation models aren’t limited to a single task. They can be adapted to various stages of the drug development pipeline, from identifying potential drug targets to designing new molecules and predicting clinical trial outcomes. This versatility is a key advantage, allowing researchers to tackle multiple challenges with a single, powerful AI engine.

Beyond Small Molecules: Targeting the Aging Brain

The collaboration isn’t just focused on traditional small molecule drugs. NVIDIA CEO Jensen Huang highlighted the potential of AI to address diseases of the aging brain, a particularly challenging area of research. Conditions like Alzheimer’s and Parkinson’s disease are characterized by complex biological processes that are difficult to model and understand.

AI can help unravel these complexities by analyzing brain imaging data, identifying biomarkers, and predicting disease progression. This could lead to the development of new therapies that target the underlying causes of these debilitating conditions.

Expanding the Ecosystem: BioNeMo and Collaborative Innovation

NVIDIA’s BioNeMo platform is playing a crucial role in democratizing access to AI-powered drug discovery tools. Recent expansions to BioNeMo include open models for RNA structure prediction and libraries for accelerating biological foundation model training. This allows researchers across the industry to leverage NVIDIA’s technology and contribute to the collective knowledge base.

The recent J.P. Morgan Healthcare Conference also saw NVIDIA recognizing leaders in the field, gifting DGX Spark systems to innovators at companies like VantAI, Boltz, and Insilico Medicine. This fosters a collaborative ecosystem, driving further advancements in AI-driven drug discovery.

The Future of Drug Manufacturing: Automation and Robotics

The impact of AI extends beyond the lab and into drug manufacturing. Companies like Multiply Labs are leveraging NVIDIA’s AI computing to automate cell therapy manufacturing, a complex and expensive process. Autonomous lab infrastructure, powered by AI, promises to increase efficiency, reduce costs, and improve the quality of manufactured drugs.

Pro Tip: Keep an eye on advancements in generative AI for molecular design. These tools are capable of creating novel molecules with desired properties, potentially leading to the discovery of breakthrough drugs.

FAQ: AI and Drug Discovery

  • What is a foundation model in drug discovery? A foundation model is a large AI model trained on a massive dataset, capable of being adapted to various downstream tasks, such as target identification and molecule design.
  • How does AI speed up drug discovery? AI accelerates the process by analyzing vast datasets, predicting drug efficacy, and automating experiments, reducing the time and cost associated with traditional methods.
  • Will AI replace human scientists? No, AI is intended to augment the abilities of scientists, not replace them. It handles complex data analysis and repetitive tasks, allowing scientists to focus on creative problem-solving.
  • What are the ethical considerations of using AI in drug discovery? Ensuring data privacy, addressing potential biases in algorithms, and maintaining transparency are crucial ethical considerations.

Looking Ahead: A New Paradigm for Healthcare

The NVIDIA-Lilly collaboration represents a pivotal moment in the evolution of drug discovery. The combination of cutting-edge AI technology, deep biological expertise, and substantial investment is poised to unlock new possibilities in healthcare. As AI continues to advance, we can expect to see even more transformative changes in the years to come, leading to faster development of more effective and personalized treatments for a wide range of diseases.

Reader Question: What role will personalized medicine play in this AI-driven future? The ability to analyze individual patient data and tailor treatments accordingly will be a key benefit of AI, leading to more effective and targeted therapies.

Want to learn more about the latest advancements in AI and healthcare? Explore our other articles or subscribe to our newsletter for regular updates.

0 comments
0 FacebookTwitterPinterestEmail
Tech

Robotics & AI Automate Cell Therapy Manufacturing for Faster, Cheaper Treatments

by Chief Editor
written by Chief Editor

The Robotic Revolution in Cell Therapy: Beyond Automation to AI-Powered Bioscience

For decades, the semiconductor industry has relentlessly pursued automation, driving down costs and increasing precision. Now, a similar transformation is underway in the world of cell therapy – a field poised to revolutionize medicine. Companies like Multiply Labs are leading the charge, swapping lab coats and manual processes for robots and sophisticated AI, promising a future where personalized medicine is not just possible, but scalable and affordable.

From Bunny Suits to Biomanufacturing Clusters: A Paradigm Shift

Traditionally, cell therapy manufacturing has been a painstakingly manual process. Scientists, often clad in full-body protective suits, meticulously handle cells, modifying them to fight diseases like cancer, genetic disorders, and autoimmune conditions. This “artisanal” approach, while effective, is incredibly expensive – often costing hundreds of thousands of dollars per patient – and vulnerable to contamination.

Multiply Labs’ approach mirrors the evolution of chip manufacturing. Instead of sterile cleanrooms reliant on human precision, they’re building controlled “biomanufacturing clusters” powered by robotics. This isn’t simply about replacing people with machines; it’s about enhancing precision, minimizing risk, and unlocking the potential for mass production. According to a recent report by Grand View Research, the global cell therapy market is projected to reach $34.99 billion by 2030, a growth rate that demands scalable manufacturing solutions.

The Power of Digital Twins and Generative AI

The real breakthrough isn’t just the robots themselves, but how they’re being trained and optimized. Multiply Labs leverages NVIDIA Omniverse to create highly detailed digital twins of their lab environments. These virtual replicas allow for risk-free experimentation and optimization of robotic workflows.

Furthermore, NVIDIA Isaac Sim is being used for imitation learning. Instead of painstakingly programming robots with every single step, scientists can simply demonstrate the desired procedure, and the AI learns to replicate it. This is crucial for capturing the “tacit knowledge” – the subtle skills and intuition – of experienced cell therapy specialists. This approach dramatically reduces development time and ensures consistent, high-quality results.

Did you know? Imitation learning can reduce robot training time by up to 80% compared to traditional programming methods.

Humanoid Robots: The Next Frontier in Lab Assistance

Multiply Labs is also exploring the use of humanoid robots, powered by NVIDIA Isaac GR00T, to assist with tasks requiring greater dexterity and adaptability. These robots aren’t intended to replace scientists entirely, but to handle repetitive or potentially hazardous tasks, freeing up human experts to focus on research and development. Imagine a robot meticulously preparing samples, monitoring cell cultures, or even assisting with complex surgical procedures – all while maintaining a sterile environment.

Beyond Manufacturing: AI-Driven Drug Discovery and Personalized Treatment

The impact of AI and robotics extends beyond manufacturing. AI algorithms are already being used to analyze vast datasets of genomic and clinical information, identifying potential drug targets and predicting patient responses to therapy. This is paving the way for truly personalized medicine, where treatments are tailored to an individual’s unique genetic makeup and disease profile.

Pro Tip: Keep an eye on companies developing AI-powered platforms for genomic analysis. These tools are likely to become essential for accelerating drug discovery and improving treatment outcomes.

Challenges and Future Outlook

Despite the immense potential, challenges remain. Regulatory hurdles, the high cost of initial investment, and the need for skilled personnel to operate and maintain these advanced systems are significant obstacles. However, as the cell therapy market continues to grow, and as AI and robotics technologies become more accessible, these challenges are likely to be overcome.

The future of cell therapy is undoubtedly automated, intelligent, and personalized. The convergence of robotics, AI, and bioscience is not just transforming manufacturing; it’s fundamentally reshaping the landscape of medicine, offering hope for patients suffering from previously incurable diseases.

FAQ

Q: How much will robotic cell therapy manufacturing reduce costs?
A: While precise figures vary, experts estimate that automation can reduce manufacturing costs by 30-50%.

Q: Will robots replace scientists in cell therapy labs?
A: No. Robots are designed to assist scientists, not replace them. They will handle repetitive and hazardous tasks, allowing scientists to focus on research and development.

Q: What is a digital twin in the context of cell therapy?
A: A digital twin is a virtual replica of a physical lab environment, used for simulation, optimization, and risk-free experimentation.

Q: What are the main benefits of using AI in cell therapy?
A: AI can improve precision, reduce contamination, accelerate drug discovery, and personalize treatment plans.

Want to learn more about the future of biotechnology? Explore our other articles or subscribe to our newsletter for the latest updates.

0 comments
0 FacebookTwitterPinterestEmail