ESMFold author: AlphaFold-style covariation could unlock small-molecule prediction
sokrypton · x · 2026-11-04
In a discussion on the huge search space of protein-ligand binding prediction, sokrypton (ESMFold author) explains his idea: AlphaFold succeeds by exploiting covariation in unlabeled protein sequences across thousands of organisms. He suggests collecting analogous data — e.g., presence/absence of small molecules per organism — might yield similar statistics for small-molecule modeling.
More from Research
- SGPnP and SNaP: making classical image reconstruction methods generative with diffusion priors — prof_kamilov · 2026-11-04
- New paper: reasoning models' CoT is far more controllable than evals suggest, usable to evade monitors — LChoshen · 2026-11-04
- BEND 2.0 benchmark: AlphaGenome tops DNA design evals, multi-objective design remains hard — Andreas Bjerregaard · 2026-11-04
- Training Humanoids in World Models via Differentiable Heightmap Lookups — breadli428 · 2026-11-04
- TScert: a sound core for TypeScript with Lean-proven checker, 628/684 functions verified — cnakazawa · 2026-11-04
- Manor Labs walks back AI self-improvement hype after closer inspection — manorlaboratory · 2026-11-04