Metraipolis › archive › dc_f445e1894f43
A System One Approach to General Protein Evolution — arXiv:2611.05396
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.05396
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-05T18:35:35.590Z
- SHA-256 of the archived text
7eac77315766dcf10ce014f2de10e67a5d0586a9f9153207eca008346bca35c7- Stable URL of this page
https://metraipolis.com/documento/dc_f445e1894f43
How to cite
A System One Approach to General Protein Evolution — arXiv:2611.05396 — arXiv (en), https://arxiv.org/abs/2611.05396 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f445e1894f43 (Metraipolis archive dc_f445e1894f43, sha256 7eac77315766dcf10ce014f2de10e67a5d0586a9f9153207eca008346bca35c7)
BibTeX
@misc{dcf445e1894f43,
title = {A System One Approach to General Protein Evolution — arXiv:2611.05396},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.05396},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f445e1894f43; sha256 7eac77315766dcf10ce014f2de10e67a5d0586a9f9153207eca008346bca35c7},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.05396. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Protein engineering requires deciding which mutation to test next and when to stop. Existing approaches often require target-family measurements, encode candidates separately, or return scores that are not calibrated to the assay. We introduce Pev, a general mutation policy that explores a System One approach to protein evolution: learning from experiments across families to make decisions in a single parallel query. Given a parent sequence and an assay context, Pev encodes the parent once and returns calibrated probabilities for choosing an edit or stopping, together with each edit's probability of improving the assay outcome. One policy covers peptides, small folded domains, and antibodies across binding and stability objectives, without fitting model weights to the target family. In ten-well screens on held-out families, Pev finds 3.3$\times$ as many beneficial mutations as random selection and 1.8$\times$ as many as zero-shot ESM-2. It also outperforms a ridge surrogate refitted on target measurements in closed-loop search on the held-out GB1 landscape. Post-hoc calibration improves probability reliability without changing decisions, while shared encoding provides over 800$\times$ the throughput of per-mutant encoding at 1,000 candidates per request. These results suggest that a System One approach could support efficient experimental decisions across protein families, with Pev providing an initial demonstration. Tema degli esploratori: scienze della vita e salute (elenco arXiv q-bio.QM, nuovi lavori del giorno). Categorie arXiv: Quantitative Methods (q-bio.QM). Pagina del lavoro: https://arxiv.org/abs/2611.05396 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-11-05T18:29:46.594Z dalla pagina https://arxiv.org/list/q-bio.QM/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).