EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Deep Prior Framework: integrating functional specificity with general plausibility for targeted protein evolution

Senxin Zhang, Yining Qin, Hanwen Zhu, Feilong Meng, Lei Jia, Xiaoqi Zheng iD

Briefings in Bioinformatics · 04.05.2026 · 27 (3)

DOI: 10.1093/bib/bbag279 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Briefings in Bioinformatics, online erschienen am 11.06.2026. Verfasst von Senxin Zhang und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Mathematics, Shanghai Normal University, No. 100 Guilin Road, Xuhui District, Shanghai 200234, China. Hauptthemen laut MEDLINE: Directed Molecular Evolution, Proteins, Computational Biology, Evolution, Molecular. Die Arbeit nennt 59 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The efficiency of directed protein evolution largely relies on computational methods to enrich mutants with high fitness. Traditional strategies, such as zero-shot approaches based on Protein Language Models (PLMs), primarily leverage general “plausibility” priors learned from natural sequences. However, in the absence of experimental feedback, their ability to guide evolution toward specific functions (“specificity”) remains limited. Here, we introduce the Deep Prior Framework (DPF), a novel paradigm that integrates universal structural plausibility with task-oriented specificity priors. DPF incorporates an innovative Bernoulli-Attention (BATT) module within a Mixture of Experts architecture, enabling efficient screening of high-fitness mutants. Benchmarking on nine deep mutational scanning datasets, DPF outperforms existing methods in terms of PLM-based method. More importantly, our method also shows high performance on in silico directed evolution of Blastobotrys adeninivorans xanthine dehydrogenase (BaXD) without intermediate experimental feedback. Experimental validation of the top-ranked mutants showed an average activity enhancement of over four-fold compared with WT, with the best mutant achieving more than a nine-fold improvement. Furthermore, we applied DPF to a large-scale annotation of unreviewed sequences in UniProt Knowledgebase (UniProtKB). Of the 42 913 366 predicted samples (~21.56% of the total), 90.07% (38 934 581 proteins) were assigned high-confidence functional labels. In summary, this study demonstrates that DPF, by incorporating specificity-aware functional priors, can significantly advance efficient and targeted protein engineering.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Directed Molecular EvolutionProteinsComputational BiologyEvolution, MolecularMutationAlgorithms

Stichwörter der Autor:innen

UniProtKBfunctional specificitymachine learningprotein evolutionprotein language models

Substanzen

Proteins

Autor:innen und Institutionen

  • Department of Mathematics, Shanghai Normal University, No. 100 Guilin Road, Xuhui District, Shanghai 200234, China.
  • Key Laboratory of RNA Innovation, Science and Engineering, Shanghai Academy of Natural Sciences (SANS), Shanghai Institute of Biochemistry and Cell Biology, Center for Excellence in Molecular Cell Science, Chinese Acade…
  • The Guangxi Key Laboratory of Intelligent Precision Medicine, 25 East Section of Gaoxin Avenue, High-tech District, Guangxi Zhuang Autonomous Region, Nanning 530007, China.
  • Department of Biomedical Informatics, School of Basic Medical Sciences, Peking University Health Science Center, No. 38 Xueyuan Road, Haidian District, Beijing 100191, China.
  • Biomaterials and Stem Cell Research Laboratory, Suzhou Institute of Biomedical Engineering and Technology, Chinese Academy of Science, No. 88 Keling Road, High-tech District, Suzhou 215011, China.
  • Center for Single-Cell Omics, School of Public Health, Shanghai Jiao Tong University School of Medicine, No. 1 Banxia Road, Pudong District, Shanghai 201318, China.

ORCID-Profile: Xiaoqi Zheng

Zitieren

APA

Zhang, S., Qin, Y., Zhu, H., Meng, F., Jia, L., & Zheng, X. (2026). Deep Prior Framework: integrating functional specificity with general plausibility for targeted protein evolution. Briefings in Bioinformatics, 27(3). https://doi.org/10.1093/bib/bbag279

Vancouver

Zhang S, Qin Y, Zhu H, Meng F, Jia L, Zheng X. Deep Prior Framework: integrating functional specificity with general plausibility for targeted protein evolution. Briefings in Bioinformatics. 2026;27(3). doi:10.1093/bib/bbag279

BibTeX
@article{Zhang2026,
  author = {Zhang, Senxin and Qin, Yining and Zhu, Hanwen and Meng, Feilong and Jia, Lei and Zheng, Xiaoqi},
  title = {Deep Prior Framework: integrating functional specificity with general plausibility for targeted protein evolution},
  journal = {Briefings in Bioinformatics},
  year = {2026},
  volume = {27},
  number = {3},
  doi = {10.1093/bib/bbag279}
}
Kostenlos starten →