News summary:
- Lilly and NVIDIA launch an AI joint innovation lab to address drug discovery challenges.
- NVIDIA is collaborating with Thermo Fisher to build an autonomous lab infrastructure for scalable scientific discovery.
- Chai Discovery, Basecamp Research, Boltz and ecosystem leaders connect NVIDIA BioNeMo, agent AI, and physical AI to scale science and drug discovery.
JP Morgan Healthcare Conference—NVIDIA today announced the following major enhancements: NVIDIA BioNeMo™ is an open development platform that enables lab-in-the-loop workflows to enable breakthrough developments in AI-driven biology and drug discovery.
The life sciences industry generates vast amounts of scientific data. BioNeMo provides a development platform for generating and processing data, training, optimizing, and deploying models. This will enable the industry to turn data into a competitive engine for discovery, maximizing the probability of success while minimizing research and development costs, currently estimated at $300 billion annually.
BioNeMo has now been expanded to include:
- new NVIDIA Clara™ open models, the RNAPro model for RNA structure prediction, and the ReaSyn v2 model to ensure the synthesis of AI-designed drugs is practical.
- BioNeMo Recipes allow you to easily accelerate and efficiently scale the training, customization, and deployment of biologically-based models.
- BioNeMo data processing libraries, including nvMolKit, a GPU-accelerated chemoinformatics tool for molecular design.
“Biology and drug discovery are at a transformative moment,” said Kimberly Powell, vice president of healthcare at NVIDIA. “BioNeMo transforms experimental data into AI-usable intelligence, so every experiment informs the next. This creates a continuous learning cycle that accelerates discovery and helps researchers build new frontier models to tackle biology's toughest challenges.”
NVIDIA is working with leading life sciences organizations to integrate BioNeMo with laboratory experiments and scientific workflows, enabling a complete AI lifecycle for biology and drug discovery and closing the loop between experimentation and AI.
today, lily announced a first-of-its-kind collaboration with NVIDIA to launch a joint innovation lab focused on tackling some of the most enduring challenges in drug discovery. At the same time, Thermo Fisher announced a partnership with NVIDIA to make scientific instruments more intelligent and increase lab autonomy.
Lilly and NVIDIA announce breakthrough joint innovation AI Lab
The NVIDIA and Lilly partnership brings together the talents of both companies, combining NVIDIA's accelerated computing, AI and robotics expertise with Lilly's world-renowned drug discovery and development proficiency. This will help Lilly pursue challenges that have the potential to revolutionize drug discovery, as the NVIDIA BioNeMo platform and Lilly's drug lab support Lilly's chemists and biologists. The companies will also explore opportunities to apply accelerated computing and advanced AI across Lilly's operations, from manufacturing to commercial operations.
The launch of this joint innovation lab follows the build-out of Lilly's NVIDIA DGX SuperPOD™ and AI factory. The most powerful biopharmaceutical. This new initiative aims to expand beyond this existing footprint and leverage investments in next-generation NVIDIA architecture, including Vera Rubin. This is expected to result in total investments in talent, infrastructure and computing of up to $1 billion over five years.
“We believe this is a catalyst for the ability to define the next era of drug discovery,” said Diogo Lau, Lilly's executive vice president and chief information and digital officer. “By working with NVIDIA, we are bringing together large-scale computing, expert talent, and the ability to shape data at massive scale. We are moving toward a future where discovery is driven by rapid experimentation and increasingly customized models. This is an approach that reflects our commitment to leading applied AI in drug discovery and making significant investments in new forms of data generation and model development.”
Thermo Fisher partners with NVIDIA to build autonomous lab infrastructure for scalable scientific discovery
By integrating NVIDIA's full-stack AI computing with Thermo Fisher's industry-leading instrumentation, our collaboration aims to transform scientific research labs into scalable, automated data factories, including:
- Integrated edge-to-cloud AI computing: Tap NVIDIA DGX Spark™ Desktop supercomputers orchestrate autonomous lab workflows and provide a seamless computing fabric from the lab edge to the cloud for high-throughput experiment management.
- Multi-agent system for lab orchestrationUse of : NVIDIA NeMo™ software suite develops agent workflows that can autonomously generate protocols, run experiments, and perform real-time quality control without the need for continuous human intervention.
- Autonomous data analysis: Integrate BioNeMo tools to autonomously interpret instrument output in real-time and accelerate the transition from raw data to actionable scientific insights.
“Artificial intelligence and laboratory automation will transform the way scientific research is conducted,” said Gianluca Petitti, executive vice president of Thermo Fisher Scientific. “Combining Thermo Fisher's leadership in laboratory technology with NVIDIA's AI solutions will help our customers work faster, improve accuracy, and get more value from each experiment, ultimately accelerating discoveries that can have significant human impact.”
AI drug discovery ecosystem powered by NVIDIA
Innovators around the world are building the future of AI for drug discovery on the BioNeMo platform, enabling developers to take an industrial-scale, AI-driven approach to understanding biology and designing potential medicines.
Biotech and drug discovery model builders who use BioNeMo to scale model training and development include:
- base camp researchintroduced the EDEN family of AI models for drug design, including a system that can precisely insert large DNA segments, a long-standing challenge in medical genetics.
- Bolts PBClaunched Boltz Lab, a software platform for AI-driven molecular design.
- chai discovery Leverage BioNeMo to accelerate the development and deployment of fundamental biomolecular models.
- Nateraannounced a proprietary AI-based modeling platform to accelerate drug discovery and design, built on proprietary and extensive genomic and clinical cancer datasets.
Apheris, Dyno Therapeutics, OpenFold, and Terray Therapeutics have also recently released models developed using the BioNeMo platform.
AI scientists build the digital lab of the future
Building agent workflows to aggregate scientific data, analyze data, generate hypotheses, and design experiments is essential to accelerating scientific discovery.
A growing ecosystem of AI scientist companies is leveraging the NVIDIA open model and NVIDIA NeMo Framework Build domain-specific agents for science, including:
- edison scientific The recently released Kosmos is an AI scientist for autonomous discovery that can perform six months' worth of work overnight.
- Tetra Science announced a partnership with Thermo Fisher Scientific to advance open and interoperable industry-aligned AI workflows, integrating NVIDIA Nemotron™ open models to extract scientific knowledge from graphs, charts, and diagrams.
- Orkin announced orkin zeroa frontier biology model built for biological discovery, trained on extensive patient data.
Other companies adopting NVIDIA NeMo and NVIDIA NIM™ microservices for AI science include Benchling, CytoReason, HelixAI (a Sapio Sciences company), and potato.
Connecting these digital agent systems to physical labs closes the loop between in silico experiments and real-world validation.
NVIDIA is working with an ecosystem of robotics and lab automation companies to deploy simulation and physical AI technologies.
- increase lab uses the NVIDIA Isaac Sim™ framework to build robot digital twins to validate and test robots before deploying them in biomanufacturing, and uses the NVIDIA Isaac GR00T model to train new operational skills.
- Lila ScienceThe Scientific Superintelligence company extends every step of the scientific method by using automated labs to generate data and validate experiments designed by AI systems.
- High resolution biosolution uses NVIDIA Isaac Sim for simulation-first lab automation design and uses NVIDIA Cosmos-Reason1 models to enable robots to interact and coordinate experiments in real-time.
- Opentrons Lab Worksa lab automation company, uses Isaac Sim to train robots to operate beyond confined environments and support AI-driven workflows that connect digital agents to physical lab operations.
Other companies such as Amgen, automatonRoche, transcription bio used the NVIDIA Omniverse™ library and Isaac Sim to build digital twins and bring physical AI to their labs and manufacturing facilities.
For more information on how it works, NVIDIA BioNeMo Platform Supporting biology and drug discovery using AI.
