EUVIMEDDescubrir EUVIMED
Open access · CC BY 4.0

Engineering selective amyloid precursor protein inhibitors by machine learning and deep mutational scanning

Reut Meiri iD, Oz Reuveni, Michal Levi, Evette S Radisky, Niv Papo iD, Yaron Orenstein iD

Protein Science · 2026-08 · 35 (8)

DOI: 10.1002/pro.70712 ↗

At a glance

Journal article in Protein Science, published online on 2026-07-20. Written by Reut Meiri and 5 co-authors. Institution: Department of Computer Science and Artificial Intelligence, Bar-Ilan University, Ramat Gan, Israel. Major MEDLINE topics: Machine Learning, Protein Engineering, Amyloid beta-Protein Precursor. 42 references. The full text is freely available (open access, CC BY 4.0). Funded by: Israel Cancer Association, NIGMS NIH HHS, NIH HHS.

ResumenSubjectsDatos bibliográficosCiteRelated studies

Abstract (original language)

Deep mutational scanning (DMS) has proven effective for mapping protein–protein interactions (PPIs), but it cannot provide complete coverage of the mutation landscape, particularly for multi‐mutant variants. To address this limitation, we trained machine‐learning (ML) models on previously generated DMS data for a stabilized amyloid precursor protein inhibitor (APPI) binding to either of two serine proteases, mesotrypsin and kallikrein‐6 (KLK6), which are implicated in various human disorders. We combined the models to accurately predict the binding selectivity of APPI variants, including double‐mutant variants, for the two serine proteases. We achieved a Pearson correlation of 0.937 between predicted log 2 selectivity enrichment ratios and DMS‐derived values. We further validated the predictions of our combined model by yeast‐surface‐display measurements and inhibition assays of purified APPI variants and revealed epistatic interactions that shape protease selectivity. Guided by binding selectivity predictions, we identified highly selective APPI variants, including the most selective mesotrypsin inhibitor reported to date. Together, these findings support the use of DMS and ML as a framework for predicting PPI selectivity and prioritizing selective therapeutic protein variants.

Subjects

MeSH terms (MEDLINE)

Machine LearningProtein EngineeringAmyloid beta-Protein PrecursorHumansMutationProtein BindingPredictive Learning Models

Author keywords

deep mutational scanningneural networksnext‐generation sequencingprotein engineeringprotein–protein interactions

Substances

Amyloid beta-Protein Precursor

Authors and institutions

  • Department of Computer Science and Artificial Intelligence, Bar-Ilan University, Ramat Gan, Israel.
  • Avram and Stella Goldstein-Goren Department of Biotechnology Engineering, Ben-Gurion University of the Negev, Beer-Sheva, Israel.
  • Department of Cancer Biology, Mayo Clinic Comprehensive Cancer Center, Jacksonville, Florida, USA.
  • National Institute of Biotechnology in the Negev, Ben-Gurion University of the Negev, Beer-Sheva, Israel.
  • The Mina and Everard Goodman Faculty of Life Sciences, Bar-Ilan University, Ramat Gan, Israel.

ORCID profiles: Reut Meiri, Niv Papo, Yaron Orenstein

Funding

  • Israel Cancer Association
  • NIGMS NIH HHS
  • NIH HHS
  • United States-Israel Binational Science Foundation
  • Rosetrees Trust
  • Israel Cancer Research Fund
  • National Institutes of Health

Cite

APA

Meiri, R., Reuveni, O., Levi, M., Radisky, E. S., Papo, N., & Orenstein, Y. (2026). Engineering selective amyloid precursor protein inhibitors by machine learning and deep mutational scanning. Protein Science, 35(8). https://doi.org/10.1002/pro.70712

Vancouver

Meiri R, Reuveni O, Levi M, Radisky ES, Papo N, Orenstein Y. Engineering selective amyloid precursor protein inhibitors by machine learning and deep mutational scanning. Protein Science. 2026;35(8). doi:10.1002/pro.70712

BibTeX
@article{Meiri2026,
  author = {Meiri, Reut and Reuveni, Oz and Levi, Michal and Radisky, Evette S and Papo, Niv and Orenstein, Yaron},
  title = {Engineering selective amyloid precursor protein inhibitors by machine learning and deep mutational scanning},
  journal = {Protein Science},
  year = {2026},
  volume = {35},
  number = {8},
  doi = {10.1002/pro.70712}
}
Start for free →