Decoding Proteins, Viruses, and Evolutionary Biology
The Complexity of Proteins
In this episode, Dmitry Korkin explores the multifaceted nature of proteins, moving beyond the simplistic view of them as linear sequences. He highlights that proteins are modular machines characterized by hierarchical complexity.
• Protein Domains: Korkin explains that proteins function as strings of protein domains, which act as both independent structural units and functional modules.
• Modular Evolution: These domains serve as the primary blocks for evolutionary innovation, allowing nature to "shuffle" capabilities while maintaining stable structural motifs.
• Dynamic Flexibility: Flexible linkers between domains play a critical role, allowing for spatial reorganization and enabling unique protein-protein interactions.
The SARS-CoV-2 Spike Protein
A significant portion of the discussion focuses on the molecular mechanics of the SARS-CoV-2 spike protein.
"There's also beautiful mechanics of how this thing happens... it's a very complex molecular process."
Korkin details how the spike protein acts as a homotrimer, with its structural dynamics—such as the opening of its receptor-binding arms—directly influencing its ability to attach to human ACE2 receptors. He suggests that targeting more stable viral components, like the M-protein lattice, may offer robust, long-term therapeutic attack vectors compared to the rapidly evolving spike protein.
Advancements in AI and Protein Folding
The conversation covers the paradigm shift in computational biology brought about by modern machine learning, specifically mentioning the impact of AlphaFold.
The Role of Expert Systems
Korkin shares a personal anecdote about his correspondence with Nobel laureate Joshua Lederberg regarding Dendral, one of the earliest expert systems. He contrasts these rule-based AI predecessors with current neural network approaches.
Machine Learning in Science
• AlphaFold's Impact: While AlphaFold represents a monumental leap in predicting the 3D structures of compact proteins, Korkin notes that understanding multi-domain, larger protein complexes remains a significant hurdle.
• Protein Design: The potential to use these systems for de novo protein engineering is immense, though it brings up necessary ethical considerations regarding the future of synthetic biology and the potential for engineered pandemics.
Historical and Personal Reflections
Korkin offers deep reflections on literature and the intersection of life and science. His recommended readings include:
• Master and Margarita by Mikhail Bulgakov
• Cancer Ward by Aleksandr Solzhenitsyn
• The Computer and the Brain by John von Neumann
• Lab Girl by Hope Jahren
The episode concludes with a poignant reading of a Russian poem, symbolizing the intersection of humanity, memory, and the cold magic of winter.