Evan Feinberg, PhD: Pioneering Molecular AI with Genesis Therapeutics
Evan Feinberg, PhD, stands at the forefront of a transformative era in drug discovery as the founder and CEO of Genesis Therapeutics. His journey began in the hallowed halls of Stanford University, where he was a graduate student in the lab of Vijay Pande, PhD. It was here that Feinberg co-invented and co-authored groundbreaking research on deep learning technologies, laying the foundation for what would become a revolutionary approach to molecular property prediction.
One of the standout innovations from Feinberg’s time at Stanford is PotentialNet, a neural network algorithm that has significantly advanced the use of graph neural networks in predicting molecular properties, particularly protein-ligand binding affinity. This work not only showcased the potential of AI in the life sciences but also led to a fruitful collaboration with Merck Research Laboratories, where Feinberg served as a deep learning consultant. This partnership validated PotentialNet’s efficacy in potency prediction, further solidifying the algorithm’s reputation in the scientific community.
Genesis Therapeutics was founded in 2019, emerging from the rich intellectual environment of Stanford. Within a year, the company secured a remarkable $52 million in Series A financing, and it has since raised over $300 million, including a substantial $200 million Series B financing round completed in 2023. The company’s rapid growth has attracted significant investment from NVentures, the venture capital arm of Nvidia, which has recognized the potential of Genesis’s innovative approach to drug discovery.
The collaboration between Genesis and Nvidia is particularly noteworthy. By leveraging Nvidia’s expertise in AI and computational efficiency, Genesis is enhancing its AI platform, the Genesis Exploration of Molecular Space (GEMS). This platform aims to generate and optimize molecules for complex biological targets by integrating advanced AI methodologies, including language models, diffusion models, and physical machine learning simulations. Feinberg emphasizes the synergy between Nvidia’s hardware and software capabilities and Genesis’s pioneering work in molecular AI, suggesting that their partnership creates a powerful combination that exceeds the sum of its parts.
A key focus of this collaboration is the optimization of equivariant neural networks, which are essential for processing 3D geometric data such as protein and small molecule structures. Nvidia’s commitment to accelerating computation through neural networks aligns perfectly with Genesis’s mission to push the boundaries of molecular AI. Feinberg notes that the field of AI is not monolithic; rather, it encompasses various subfields that utilize distinct algorithms tailored to specific tasks, such as drug discovery.
Feinberg’s academic background at Stanford, particularly his work with Pande, has been instrumental in shaping his vision for Genesis. Pande, who has transitioned to a general partner at Andreessen Horowitz, played a crucial role in securing initial funding for Genesis and continues to influence the company’s strategic direction. Feinberg reflects on the invaluable mentorship he received from Pande, highlighting how their collaborative efforts have propelled Genesis to the forefront of molecular AI.
As the field of AI has evolved, so too has Feinberg’s approach to drug discovery. During his graduate studies, AI’s impact was primarily felt in computer vision and natural language processing. Recognizing the limitations of existing neural network types for chemistry, Feinberg and his colleagues developed new algorithms better suited for molecular applications. This innovative spirit continues to drive Genesis, as the company consistently explores new AI algorithms and neural network architectures tailored for molecular AI.
The GEMS platform is designed to tackle some of the most challenging targets in drug discovery, aiming to generate molecules with optimal potency, selectivity, and atomy characteristics. GEMS integrates generative and predictive AI, allowing Genesis to produce vast libraries of compounds while ensuring that predictive models accurately assess their potential efficacy. This multi-parameter optimization approach is crucial for advancing drug candidates through the development pipeline.
Genesis is currently focused on oncology and immunology, with several promising projects underway. In oncology, the company is nearing the nomination of highly potent and selective development candidates for pan-mutant allosteric inhibitors of PIK3CA, a key oncogenic driver in breast and colorectal cancers. Additionally, Genesis is working on small molecules that aim to enhance responses to checkpoint inhibitors and prevent cancer cell evasion of apoptosis.
In the realm of immunology, Genesis is developing treatments for autoimmune disorders and a severe genetic autoinflammatory disease. The company has also established collaborations with major biopharma players, including Gilead Sciences, Eli Lilly, and Genentech, to leverage its GEMS platform for drug discovery across multiple targets. These partnerships not only provide financial backing but also enhance Genesis’s capabilities in developing innovative therapies.
Headquartered in Burlingame, California, with a fully integrated laboratory in San Diego, Genesis Therapeutics employs around 80 people and is poised for significant growth. Feinberg anticipates expanding the team in response to the influx of capital from the Series B financing and ongoing partnerships, reflecting the company’s ambitious plans for the future.
As Evan Feinberg continues to lead Genesis Therapeutics, his vision for integrating AI into drug discovery is reshaping the landscape of the life sciences. With a commitment to innovation and collaboration, Feinberg and his team are well-positioned to unlock new possibilities in the quest for effective therapies, ultimately improving patient outcomes and advancing the field of molecular AI.
