Evan Feinberg, founder and CEO at Genesis Therapeutics, talks about using AI in drug discovery and the collaboration between ML, biotech, and chemistry. They discuss how physics-based models can improve drug design, integrating human intuition and predictive AI models, and the importance of a multidisciplinary team in drug discovery programs.
Genesis Therapeutics uses AI to tackle drug discovery through their interdisciplinary approach, combining chemists, software engineers, and machine learning researchers to develop their internal platform called Nucleus.
AI in drug discovery requires a combination of human expertise and machine learning algorithms specifically tailored to the complex world of drug design, as the field necessitates extrapolation and the ability to discover new chemical matter.
Deep dives
Genesis Therapeutics: Tackling Drug Discovery through AI
Evan Feinberg, founder and CEO of Genesis Therapeutics, discusses how his work in the lab led to the creation of Genesis, which uses AI to tackle the problem of drug discovery. The company takes an interdisciplinary approach, combining the expertise of chemists, software engineers, and machine learning researchers to develop their internal platform called Nucleus. This platform allows chemists to access the latest ML models for predicting various properties, enhancing productivity and providing human feedback. Feinberg emphasizes the importance of unifying biotech with AI to fundamentally change the drug discovery process.
The Counterintuitive Nature of AI in Life Sciences
Feinberg highlights the counterintuitive nature of using AI in life sciences, comparing it to the shift from Newtonian to quantum physics in the 20th century. He explains that while image and text data have abundant training data, the field of drug design requires extrapolation and the ability to discover new chemical matter. The field of computational chemistry has immense promise, but deep learning algorithms specifically tailored to drug discovery were lacking until recent advancements. Feinberg emphasizes the need for human expertise in collaboration with AI to develop models that make sense in the complex world of drug discovery.
Building Generative AI and Predictive AI for Drug Design
Genesis Therapeutics focuses on two main technological pillars: generative AI for chemistry and predictive AI. Their generative AI models produce compounds that are drug-like by deeply integrating human intuition and drug likeness into the training process. The predictive AI component allows for predicting how ligands will interact with protein targets and assessing potency and selectivity. The company has developed dynamic potential net, a technology that combines physics-based simulation and deep neural networks, merging two previously distinct pillars. Genesis aims to be as vertically integrated as possible and collaborates with academic institutions and large pharmaceutical companies to tackle challenging drug targets.
Evan Feinberg, PhD, founder and CEO at Genesis Therapeutics, joins Vijay Pande of Bio + Health.
Together, they talk about how Evan’s work in the lab (ironically, Vijay's lab at Stanford!) translated to the creation of Genesis, which is tackling the problem of drug discovery through AI. They also discuss how the Genesis team is building specifically to carry on work at the intersection of ML, biotech, and chemistry.
Get the Snipd podcast app
Unlock the knowledge in podcasts with the podcast player of the future.
AI-powered podcast player
Listen to all your favourite podcasts with AI-powered features
Discover highlights
Listen to the best highlights from the podcasts you love and dive into the full episode
Save any moment
Hear something you like? Tap your headphones to save it with AI-generated key takeaways
Share & Export
Send highlights to Twitter, WhatsApp or export them to Notion, Readwise & more
AI-powered podcast player
Listen to all your favourite podcasts with AI-powered features
Discover highlights
Listen to the best highlights from the podcasts you love and dive into the full episode