Dmitry Korkin: Evolution of Proteins, Viruses, Life, and AI | Lex Fridman Podcast #153

In this episode, Dmitry Korkin, a professor of bioinformatics and computational biology at WPI, explores the fundamental nature of proteins, viruses, evolution, and the revolutionary impact of artificial intelligence on biological science. The conversation begins with an accessible explanation of proteins as the building blocks of life. Korkin describes how proteins fold into intricate three-dimensional structures that determine their function, using the example of the spike protein found on coronaviruses. This spike protein is particularly important because it acts as the key that allows the virus to enter human cells, making it a critical target for vaccines and therapeutic interventions.

The discussion then moves to the coronavirus structure itself, examining how the virus is organized at a biological level and how mutations in the viral genome lead to new variants. Korkin explains that viruses evolve through the same mechanisms as all life forms, acquiring mutations that sometimes provide advantages in evading the immune system or infecting new hosts. This naturally leads to a broader conversation about protein evolution more generally, exploring how proteins have changed and diversified over billions of years to create the vast complexity of life.

A particularly fascinating segment explores self-replicating computer programs as an analogy for understanding how life might have originated. This conceptual bridge leads to deeper questions about the origin of life itself, the conditions required for self-replication to begin, and whether life might exist elsewhere in the solar system on places like Europa or Enceladus. Korkin references Joshua Lederberg and early AI systems like Dendral, reflecting on why expert systems ultimately failed despite initial promise.

The conversation then pivots to modern artificial intelligence, specifically AlphaFold 2, which represents perhaps the most significant breakthrough in computational biology in decades. AlphaFold can predict the three-dimensional structure of proteins from their amino acid sequences with unprecedented accuracy, solving a problem that has challenged scientists for fifty years. Korkin discusses the implications of this technology for drug discovery, disease understanding, and biomedical research. The episode explores whether AlphaFold's developers will receive a Nobel Prize for this achievement and whether AI will revolutionize creative fields like art and music.

An important section addresses the ethical implications of advanced AI in biology. Korkin candidly discusses concerns about whether AI could be used to engineer deadly viruses or create biological weapons. This sobering topic highlights the dual-use dilemma that increasingly confronts powerful scientific and technological tools. The episode concludes with book recommendations, personal reflections on family, and Korkin reciting a poem in Russian, providing a more humanistic end note to the technical discussion.