Claude MSA: full pipeline recalling homologs from memory, 43-56% top-L contact accuracy
sokrypton · x · 2026-11-20
The full thread with GitHub gist "recallmsa.py": a single-file pipeline from prompt to contacts, where Claude Fable 5.2 generates 200 distant homologs per lineage group purely from memory with no retrieval tools. Coevolution contact maps for YaaA, CheY, thioredoxin and DHFR show 43–56% top-L pair accuracy (e.g. YaaA 110/256, DHFR 87/159), with clearly readable coevolution signal — the most complete public validation yet that LLMs internalize protein family distributions. Script and $15/MSA cost included.
More from Research
- Image Models Now Ship Fine-tuned LLM Prompt Upsamplers — Why Not Fuse Them? — ostrisai · 2026-11-20
- PyTorch's ezyang on formalizing a trivial in-place kernel under PTX semantics — ezyang · 2026-11-20
- Scientist: Gemini 4 Neon designed innovative CAR receptors to kill cancer cells — DeryaTR_ · 2026-11-20
- New arXiv paper: Bregman power divergences turn KL queries into Euclidean NN in d+1 — FrnkNlsn · 2026-11-20
- Untrained Neural Networks Hit 86.4% AUROC on ADBench Anomaly Detection — ideal-lab-sejong · 2026-11-20
- NAVER's Unfinished-Bench: Persistent Agents Break Hidden Rules 43.6% of the Time — naver-ai · 2026-11-20