Generative AI Applications in Drug Discovery
As a part of the Roche Group, Genentech is leading the way in the application of generative AI to find and create novel therapies and improve patient care.Through the optimization and acceleration of Genentech’s proprietary algorithms, a new partnership between NVIDIA and the biotechnology pioneer Genentech seeks to revolutionize the discovery and development of new medications by bringing together professionals from both companies.
In order to expedite these models on NVIDIA DGX Cloudwhich offers specialized AI supercomputing instances and software hosted by NVIDIA cloud service provider partnersNVIDIA and Genentech will collaborate.
Genentech intends to employ NVIDIA BioNeMo, which enables biotech businesses to seamlessly integrate BioNeMo cloud application programming interfaces into computational drug development workflows, and tailor models at scale.
BioNeMo is a domain-specific platform that streamlines, speeds up, and scales generative AI applications for computational drug development. It is currently offered as a training service. On DGX Cloud, it enables researchers to pretrain or optimize cutting-edge models.
Initially, the partnership will concentrate on improving Genentech’s AI models for drug development within its “lab in a loop” structure. The objective is to enable scientists to produce multiplicative, as opposed to linear or additive, benefits for patients and the larger healthcare ecosystem. This will be accomplished by enabling researchers to comprehend intricate biomolecular patterns and relationships, which will genuinely disrupt drug development and increase the success rate of R&D.
“We have a long history of successfully developing and implementing technology in ways that weren’t immediately obvious to others,” stated Aviv Regev, executive vice president and head of Genentech Research & Early Development (gRED), on their partnership with NVIDIA. “We revolutionized the pharmaceutical industry by being the pioneer biotech business to utilize molecular biology in medication research and development.
We invented antibody therapies, which later on emerged as the standard of care. Currently, we are combining AI, the lab, and the clinic to find patterns in massive amounts of data that would not have been possible to access otherwise, and to create studies to test those patterns. Working with NVIDIA and introducing generative AI has the potential to accelerate the search for and development of treatments that will enhance the quality of life for patients everywhere.
Currently, the process of finding and developing new drugs is drawn out, difficult, and expensive. It is challenging to forecast drug targets for new medications, just as it is challenging to successfully create a molecule as a possible therapy. Because generative and other AI models can assist scientists in quickly identifying possible therapeutic compounds and interactions through training on massive datasets, AI has the potential to be a transformative technology.
Using AI, according to Genentech, helps close the knowledge gap between computational algorithms and lab research. The company’s R&D team, gRED, has already made great progress in applying AI, across a variety of modalities to find and create new treatments while learning more about the fundamentals of biology and illness.Teams from Genentech and NVIDIA will now collaborate to optimize Genentech’s proprietary models in an effort to reduce the length of this laborious drug discovery and development process and increase its success.
Including AI in a Cycle
The “lab in a loop” tool from Genentech provides an iterative framework for creating and examining molecular designs with anticipated characteristics. Its goal is to improve the optimization of next molecular designs by using experimental data to inform generative computational models. NVIDIA will speed up the training and inference of Genentech’s drug discovery models, assisting Genentech in optimizing its framework.
NVIDIA AI experts will learn more about AI-related difficulties in medication research and discovery through our collaboration. With these insights, NVIDIA intends to enhance its BioNeMo platform and other products to better meet the needs of models utilized by the biotech sector.
Vice President of Healthcare at NVIDIA Kimberly Powell stated, “AI can play a transformational role in accelerating drug discovery and development as it has across many parts of healthcare and life sciences.” In order to quickly iterate and uncover insights, NVIDIA and Genentech are working together to build and apply AI models and algorithms that are unlocking scientific discovery.