The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence) Open-Source Drug Discovery with DeepChem with Bharath Ramsundar - #566
11 snips
Apr 4, 2022 
Bharath Ramsundar, founder and CEO of Deep Forest Sciences, shares his expertise in AI-driven drug discovery and molecular design. He delves into the challenges biotech firms face in integrating AI, highlighting the need for collaboration and a solid infrastructure. The discussion includes the innovative DeepChem library and its datasets like MoleculeNet, which aim to enhance drug development processes. Bharath also emphasizes the importance of chemistry-aware validation methods for better model generalization and the evolving partnership between AI and traditional sciences.
AI Snips
Chapters
Books
Transcript
Episode notes
Challenges of AI in Drug Discovery
- Biotech and pharma companies face challenges incorporating AI due to IT infrastructure, pay gaps, and talent acquisition.
- Effectively integrating AI with human scientists requires clear communication and bridging computational insights with biology/chemistry.
AI's Promise in Drug Discovery
- Machine learning for chemistry has advanced significantly, particularly in predicting molecular properties.
- Machine learning for biology is less mature, except for AlphaFold 2, and faces the challenge of understanding complex biological systems like diseases.
DeepChem's Origin
- Bharat Ramsundar's internship at Google used an early AI system called Disbelief, pre-TensorFlow.
- After the internship, he replicated the work using Keras, shared it on GitHub, and it grew into DeepChem.
