EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Artificial intelligence for protein engineering: From prediction and generation to evidence-centred design cycles

Yugeng Liu, Heng Wang, Xiaojian Zhang, Guihong Cha

Artificial Intelligence in the Life Sciences · 2026-11 · 100193

DOI: 10.1016/j.ailsci.2026.100193 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Artificial Intelligence in the Life Sciences, 2026-11. Verfasst von Yugeng Liu und 3 weiteren Autor:innen. Die Arbeit nennt 117 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Artificial intelligence now supports mutation ranking, structure modelling, inverse folding, generative design and experiment selection in protein engineering. These tools can propose far more candidates than laboratories can test. This critical review examines data resources and benchmarks, computational roles and objective-specific workflows, and appraises the quality and limits of the experimental evidence supporting their use. It covers enzymes, binders, stability and developability, therapeutic and membrane proteins, assemblies, biosensors and genome editors. Across these applications, structural self-consistency, model likelihood and docking scores can help prioritise candidates but do not establish function. Prospective experimental success is distinguished from demonstrated improvement in candidate selection, which requires relevant comparators. We apply six domains to the prospective campaigns surveyed here: objective–assay alignment; data provenance and generalisation; calibrated computation and selection; prospective and orthogonal validation; reproducibility and traceability; and governance and translation. We recommend coupling context-aware models to experiments that test their hypotheses and inform the next design decision.

Zitieren

APA

Liu, Y., Wang, H., Zhang, X., & Cha, G. (2026). Artificial intelligence for protein engineering: From prediction and generation to evidence-centred design cycles. Artificial Intelligence in the Life Sciences, 100193. https://doi.org/10.1016/j.ailsci.2026.100193

Vancouver

Liu Y, Wang H, Zhang X, Cha G. Artificial intelligence for protein engineering: From prediction and generation to evidence-centred design cycles. Artificial Intelligence in the Life Sciences. 2026:100193. doi:10.1016/j.ailsci.2026.100193

BibTeX
@article{Liu2026,
  author = {Liu, Yugeng and Wang, Heng and Zhang, Xiaojian and Cha, Guihong},
  title = {Artificial intelligence for protein engineering: From prediction and generation to evidence-centred design cycles},
  journal = {Artificial Intelligence in the Life Sciences},
  year = {2026},
  pages = {100193},
  doi = {10.1016/j.ailsci.2026.100193}
}
Kostenlos starten →