Genesis to work with NVIDIA to optimize its neural network architectures around 3D geometric data of proteins and drug-like molecules.
Artificial intelligence-powered drug discovery company Genesis Therapeutics has announced a new equity investment and strategic collaboration with tech giant NVIDIA, aiming to push forward advancements in its AI platform. The equity contribution comes from NVentures, the venture capital division of NVIDIA, which also took part in the last year’s oversubscribed $200 million Series B round and brings the total raised by the company to more than $300 million to date.
Based in Burlingame, CA, Genesis applies advanced AI methods to identify and target “undruggable” proteins. Leveraging its GEMS AI platform, Genesis combines generative and predictive AI models to accelerate the development of small molecule drugs, addressing complex therapeutic targets with unprecedented precision. The platform integrates AI techniques such as language and diffusion models with physical machine learning simulations, creating a powerful system to generate and optimize molecules.
“Addressing previously undruggable targets to create new medicines requires pioneering algorithms for drug discovery and fundamental infrastructure innovation,” said Dr Evan Feinberg, founder and CEO of Genesis. “Working with NVIDIA, we will further the capabilities of our field-leading physical AI platform for structure-driven drug design.”
A significant aspect of the GEMS platform is its ability to understand and interpret the three-dimensional structure and interactions between proteins and potential drug molecules. Using a graph-based approach to represent protein-ligand complexes, the platform makes accurate predictions based solely on 3D structural information. This understanding is crucial for designing effective drugs, especially for targets considered difficult or impossible to address using traditional methods. GEMS also features a molecular simulation component that captures 3D binding dynamics, enhancing its ability to generate innovative molecules that can be explored in specific chemical spaces tailored to each drug development program.
The collaboration with NVIDIA aims to bring new computational efficiencies to Genesis’s platform, particularly through the optimization of neural network architectures critical for drug discovery. One of the main areas of focus will be refining equivariant neural networks, which are essential for managing 3D geometric data, such as the structures of proteins and drug-like molecules. By making these computations more efficient, Genesis and NVIDIA aim to improve the overall performance of the company’s AI models, expediting the identification of promising drug candidates and enhancing molecular design capabilities.
“AI serves as a tool for discovering new therapies to treat diseases,” said Mohamed “Sid” Siddeek, corporate vice president at NVIDIA and head of NVentures. “Genesis’ AI platform, and related computational advancements developed in collaboration with NVIDIA, will help deliver novel generative and predictive AI techniques to explore untapped chemical pathways and identify drug candidates.”
In addition to collaborations with pharma industry giants Genentech and Eli Lilly, Genesis also recently announced a strategic collaboration with Gilead Sciences to discover and develop novel, small molecule therapies across multiple targets.
