Bristol Myers Squibb’s Leap into AI: Acquiring Nvidia’s DGX SuperPOD for Drug Discovery
Bristol Myers Squibb (BMS), a leading name in the pharmaceutical industry, is taking a transformative step in its research and development capabilities by acquiring an Nvidia DGX SuperPOD. This powerful AI infrastructure is based on Nvidia’s new Vera Rubin architecture. The move marks BMS as the first life sciences entity to embrace this advanced AI system, introduced by Nvidia earlier this year, signaling a major leap in the integration of artificial intelligence in drug discovery and development.
Expansion of Computing Capacity
The newly acquired DGX SuperPOD will comprise eight DGX Vera Rubin NVL72 systems. Each system is equipped with cutting-edge Nvidia Vera CPUs and Rubin GPUs, designed to significantly enhance BMS’s computational capabilities. This infrastructure will facilitate the training of proprietary models and enable complex predictions across BMS’s vast array of research programs, covering compounds, proteins, and other crucial scientific data.
While financial specifics of the acquisition remain undisclosed, this purchase builds on BMS’s existing Nvidia infrastructure, which includes an older SuperPOD. This older system, described by BMS executives as two to three generations behind the Vera Rubin architecture, has been operational for about three years. The new system will be integrated with the existing infrastructure, creating a unified computing environment accessible to BMS’s research sites globally.
The SuperPOD’s software stack is designed to efficiently schedule training, prediction, and development workloads, thereby providing more scientists with direct access to substantial computing resources. Greg Meyers, BMS’s chief digital and technology officer, emphasized the growing computational demands due to the deployment of larger AI models within the organization.
Application of AI in Drug Research
BMS has been pioneering the application of AI across its research initiatives. AI now influences every small molecule program and the majority of large molecule programs at BMS, assisting in target identification, lead optimization, and large molecule predictions. The technology has proven instrumental in reducing manual research tasks, cutting down timeframes significantly.
Dr. Robert Plenge, BMS’s Chief Research Officer, highlighted the transformative impact of AI on drug candidate evaluation, allowing researchers to assess more potential treatments during early development stages. This efficiency is achieved through a strategy known as “Predict First,” which uses AI to pre-screen molecules, ensuring only the most promising candidates proceed to synthesis and laboratory testing.
AI has also been harnessed to expand BMS’s CELMoD compound library, targeting cancer-causing proteins with precision. This approach has accelerated the exploration of new protein targets and potential therapeutic compounds, thereby enhancing the overall drug discovery process.
The integration of Nvidia’s BioNeMo Agent Toolkit with the Vera Rubin system will further empower BMS researchers. BioNeMo offers advanced tools for protein structure prediction, molecule generation, and sequence analysis, enhancing the research workflow significantly.
Connecting Research Sites
BMS is committed to democratizing access to its advanced computational resources. The company is implementing user-friendly interfaces that allow researchers to initiate complex computational tasks using natural language commands. The environment is managed via Nvidia Mission Control, offering streamlined cluster provisioning, infrastructure monitoring, and workload management.
This unified infrastructure enables seamless data sharing across BMS’s global research network. For instance, data generated in Lawrenceville, New Jersey, can be utilized by teams in San Diego, ensuring a cohesive and integrated research approach. This setup not only preserves valuable experimental data but also institutionalizes research findings across BMS’s operations.
The new computing capacity will support various research areas, including molecule design, clinical research, and digital twin applications. While specific details on the deployment timeline and hosting site for the new system remain under wraps, the enhanced computational efficiency, delivering up to 10 times more performance per megawatt compared to older systems, is a significant advantage.
As BMS continues to push the boundaries of AI application in pharmaceuticals, this strategic acquisition underscores its commitment to innovation and excellence in drug discovery and development.
Want to learn more about AI and big data from industry leaders? Check out the AI & Big Data Expo taking place in Amsterdam, California and London. The comprehensive event is part of TechEx and takes place alongside other leading technology events including the Cyber Security & Cloud Expo. Click here for more information.
AI News is operated by TechForge Media. Discover more upcoming enterprise technology events and webinars here.
“`
