Advancing Therapies for Immune Diseases with Kfir Schreiber from DeepCure

Can AI cure autoimmune diseases? This episode of Impact AI dives into the groundbreaking work of DeepCure, where artificial intelligence meets medicinal chemistry to tackle some of healthcare's most stubborn challenges. Co-founder and CEO Kfir Schreiber shares how his team uses advanced machine learning tools, physics simulations, and human expertise to design the next generation of small molecule drugs. From overcoming data limitations to fostering tight collaboration between machine learning scientists and chemists, this discussion illuminates the potential of AI-driven innovation in transforming patient outcomes. With a rheumatoid arthritis drug nearing clinical trials, DeepCure is poised to redefine the future of medicine. Tune in to discover how AI can accelerate drug discovery, overcome data challenges, and create life-changing therapies, as well as how these insights can inspire your own innovative pursuits!

Key Points:

• How Kfir's background in computer science and applied math led him to found DeepCure.
• Insight into DeepCure’s mission to leverage proprietary technology to create small molecule drugs for inflammation and autoimmunity.
• Augmenting human expertise with AI: the role of machine learning in drug discovery.
• Layers of using AI to analyze targets and design small molecules with optimized properties.
• Challenges in small molecule datasets and how DeepCure develops tailored models.
• The influence of molecule representations like SMILES on machine learning models.
• Combining publicly available datasets with data generated in DeepCure’s automation lab.
• Model validation techniques to address out-of-distribution challenges in small molecule data.
• Collaboration between machine learning experts and chemists to refine drug discovery.
• Recruiting top talent by highlighting DeepCure’s impactful mission in healthcare.
• The process of onboarding machine learning developers with no prior chemistry knowledge.
• Problem-solving advice for leaders of AI-powered startups: it’s not about the AI!
• DeepCure’s future plans for clinical trials and expansion into other autoimmune diseases.

Quotes:

“Machine learning in our space is almost never a complete solution. It's a way to augment our chemists [and] our biologists [to] try to make them capable of solving problems that were unsolved before.” — Kfir Schreiber

“One of the best things about DeepCure [is the] very tight collaboration between the domain experts and our machine learning scientists.” — Kfir Schreiber

“Your average machine-learning scientist doesn't have chemistry intuition. We need this feedback and we need to integrate this feedback back into our models to make the predictions make sense.” — Kfir Schreiber

“Focus on the problem, focus on the value, and work your way backwards to the best tools to use.” — Kfir Schreiber

Links:

DeepCure
Kfir Schreiber on LinkedIn

Resources for Computer Vision Teams:

LinkedIn – Connect with Heather.

Computer Vision Insights Newsletter – A biweekly newsletter to help bring the latest machine learning and computer vision research to applications in people and planetary health.

Computer Vision Strategy Session – Not sure how to advance your computer vision project? Get unstuck with a clear set of next steps. Schedule a 1 hour strategy session now to advance your project.