Covariation made AlphaFold and LLMs possible — small molecules need equivalent data
proteinrosh · x · 2026-11-04
sokrypton and proteinrosh debate how small-molecule AI could replicate AlphaFold's success:
- sokrypton: protein breakthroughs came from covariation data enabled by genomics/metagenomics; for small molecules we need ways to identify ligand presence/absence across the whole tree of life.
- proteinrosh: sequencing is cheap and informative on its own — coevolution data just happened to become cheap to collect via sequencing.
- sokrypton: covariation is the key ingredient (for LLMs too), so other modalities need equivalent data collected under natural or artificial selection — before biodiversity goes extinct.
More from Research
- 2-person lab Linum trains 2B text-to-image model from scratch for just $21K, Apache 2.0 — ycombinator · 2026-11-05
- Trained an MoE across two GPU clusters 25 Gbps apart — the compute islands guide — _arohan_ · 2026-11-05
- AutoDNA: AI agents run multi-step molecular biology experiments with minimal human input — james_y_zou · 2026-11-05
- The Stacks Project, algebraic geometry's giant reference, is now formalized in Lean — LinusEkenstam · 2026-11-05
- Study benchmarks divergent creativity of LLMs against 100,000 humans in Scientific Reports — korymath · 2026-11-05
- Quanta: injecting "thoughts" into LLMs to test if models can introspect — coherence · 2026-11-05