EvoMax Evolves Compact Genome Editor to 97% Efficiency Using Sparse Data
bravo_abad · x · 2026-08-27
Addressing the data scarcity problem in protein engineering, Shijie Wan et al. introduce EvoMax, a model-guided strategy designed for the sparse-data regime. They used it to engineer compact eukaryotic genome editors achieving 97% endogenous editing efficiency.
EvoMax combines three complementary signals: a Gaussian Process Regression (GPR) learning directly from measured mutational fitness; ESM-2 contributing evolutionary information; and ESM-IF adding structure-conditioned inverse-folding info. The relative contributions shift across optimization rounds, allowing the search to move from local exploitation to broader exploration. The initial GPR uses only 209 experimentally measured single-site mutations.
More from Research
- CrysVCD enforces chemistry constraints in crystal generation — bravo_abad · 2026-08-27
- Continual Learning is AI's Holy Grail: Fluid intelligence will enable the next capability leap — daniel_mac8 · 2026-08-27
- How to achieve same intelligence with less computation? Exploring 7 technical paths — prateekj · 2026-08-27
- AI-designed peptide guides enable mammalian protein screens — AllThingsApx · 2026-08-27
- Prime Intellect Releases Technical Report for Prime Agent Framework — xeophon · 2026-08-27
- Where LLM Document Audits Fail in Production: Compliance & Finance — gnatykdm · 2026-08-27