a16z Podcast a16z Podcast: Deep Learning for the Life Sciences
Jun 6, 2019
Vijay Pande, a General Partner at A16Z and co-author of "Deep Learning for the Life Sciences," teams up with Bharath Ramsundar to explore how AI is transforming biology. They discuss the surge of AI applications in genomics and drug discovery, noting the rise of open-source practices that democratize research. The duo emphasizes the importance of collaboration between data scientists and biologists, challenges in AI integration, and the potential for a revolutionary 'open source biology' movement, ultimately reshaping how we approach life sciences.
AI Snips
Chapters
Books
Transcript
Episode notes
Deep Learning's Impact on Life Sciences
- Deep learning's recent impact on life sciences is due to a confluence of factors, not just one breakthrough.
- These factors include improved hardware and software, open-source code sharing, and the ability to learn complex representations from raw biological data.
Importance of Data Representation
- Choosing the right data representation is crucial for effective computation, just like converting Roman numerals to Arabic numerals for addition.
- Molecules, unlike images or text, lack obvious representations, making their computational analysis challenging.
Maturing of Biological Deep Learning
- Deep learning in biology is maturing beyond simply applying image-based techniques.
- It's developing its own methods, like graph deep learning for molecules, diverging from image or text-based approaches.
