a16z Podcast Journal Club: Finding New Antibiotics with Machine Learning, What Coronavirus Structures Tell Us
Apr 26, 2020
Vijay Pande, a general partner at Andreessen Horowitz with a focus on biotechnologies, joins Judy Savitskaya, a bio deal partner specializing in virology. They delve into groundbreaking machine learning methods for antibiotic discovery, particularly the novel antibiotic Halicin. The conversation reveals the potential to combat antibiotic resistance. Savitskaya also shares insights on the coronavirus structures, linking them to past SARS research and discussing their implications for vaccine development and viral entry mechanisms.
AI Snips
Chapters
Books
Transcript
Episode notes
Antibiotic Development Challenges
- Scientific challenges and business obstacles hinder antibiotic development.
- Innovative antibiotics are reserved for severe cases, limiting their profitability.
Deep Learning for Antibiotic Discovery
- Deep learning models can predict antibiotic effectiveness without preset chemical structure info.
- This allows for identifying novel drug structures, unlike traditional methods that yield similar molecules.
Halicin's Properties
- Halicin disrupts bacterial pH balance, a novel mechanism.
- It effectively inhibits various pathogens, including E. coli and C. diff, and eliminates persistent cells.
