AlphaFold Database adds predicted structures for 2,800+ viruses, 30% interactions undocumented
AllThingsApx · x · 2026-10-01
An international team (EMBL-EBI, Google DeepMind, NVIDIA, CEPI, Seoul National University, and others) ran 41,774 viral proteins through AlphaFold2 optimized with NVIDIA BioNeMo, generating 1.7M predicted pairings. About 8,000 passed quality filtering into a curated, freely available database, with raw predictions also released. Roughly 30% of added interactions had never been documented. Viruses are a structural biology blind spot since many express polyproteins cut into pieces, making protein boundaries hard to infer from genomes alone.
More from Research
- METR time-horizon chart updated: doubling every 1.7 months since Opus 4.5, 'Model 4' ~38h — ben_j_todd · 2026-11-16
- GTAgentClinic: one-line install, plug in Claude Code or Codex for a 3D hospital shift — Michael_D_Moor · 2026-11-16
- 28-model tracking table: half publish no training cutoff, Gemini 4 Carbon 8 months stale at launch — Joozio · 2026-11-16
- FutureSim Q3: LLMs now genuinely improve from experience, Opus 5.5 tops at 50% — rickasaurus · 2026-11-16
- Dataset Distillation via Drifting Compresses ImageNet Classes to Just 10 Images — giannis_daras · 2026-11-16
- Why divide by n-1: a 2-minute explainer on Bessel's correction — JFPuget · 2026-11-16