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

Korkin emphasizes that most proteins are composed of multiple domains—modular structural and functional units that get reused, shuffled, and recombined across evolution, making domains a more meaningful 'building block' than entire proteins.

While the spike trimer and its receptor-binding domains mediate entry via ACE2, the more evolutionarily stable membrane (M) protein forms a lattice that organizes the viral envelope and may be a promising, less mutation-prone target for small‑molecule drugs.

Mutations enable viruses to adapt, cross species, and potentially evade vaccines or treatments; tracking sequence changes across geography and hosts, and modeling their functional impact, may let us forecast which strains or mutations are likely to become problematic.

It achieves near‑experimental accuracy for many single‑domain or compact proteins in CASP benchmarks, yet multi‑domain, highly flexible proteins and multi‑protein complexes remain unsolved, and the fundamental physical mechanism of folding is still not understood.

Korkin notes that successful systems like AlphaFold embed detailed biological priors (evolutionary relationships, structural constraints), showing that raw deep learning alone is not enough; structured domain knowledge still drives major gains.

The same models that predict pathogenicity or structural effects of mutations to aid pandemic preparedness could, in principle, be misused to suggest more dangerous variants—highlighting the need for regulation, transparency, and careful governance.

Compared to SARS, the structural characterization and sequencing of SARS‑CoV‑2 have happened in months instead of years, enabling rapid vaccine design and detailed evolutionary tracking, and illustrating how global scientific collaboration can accelerate under pressure.