EUVIMEDScopri EUVIMED
Open access · CC BY-NC-ND 4.0

Supervised learning of protein variant effects across large‐scale mutagenesis datasets

Thea K Schulze, Lasse M Blaabjerg, Matteo Cagiada, Kresten Lindorff-Larsen iD

Protein Science · 2026-04 · 35 (4)

DOI: 10.1002/pro.70526 ↗

At a glance

Journal article in Protein Science, published online on 2026-03-07. Written by Thea K Schulze and 3 co-authors. Institution: The Linderstrøm-Lang Centre for Protein Science, Department of Biology, University of Copenhagen, Copenhagen, Denmark. Major MEDLINE topics: Proteins, Supervised Machine Learning, Mutagenesis. Cited 2 times so far (Crossref); 85 references. The full text is freely available (open access, CC BY-NC-ND 4.0). Funded by: Novo Nordisk Fonden.

AbstractSubjectsDati bibliograficiCiteRelated studies

Abstract (original language)

The increasing availability of data from multiplexed assays of variant effects (MAVEs) enables supervised model training against large quantities of experimental data to learn sequence‐function relationships. Variant effect scores from MAVEs can, however, be influenced by the experimental method and library composition, resulting in experiment‐to‐experiment differences in the mapping from molecular level variant effects to MAVE readout. This variation presents a challenge for supervised learning across datasets that limits our ability to leverage the data generated by MAVEs. We here develop a framework for performing supervised learning with MAVE data that takes the influence of the experimental protocol into account, thus enabling variant effects to be learned across datasets produced in independent experiments. We apply the framework to train a model against variant effect scores collected with VAMP‐seq, a MAVE technique that quantifies the steady‐state cellular abundance of protein variants. Our results suggest that mapping variant abundance to VAMP‐seq readout in a dataset‐specific manner during model training improves the learned abundance model and moreover allows the learned model to predict variant effects on an interpretable scale. Our work highlights the importance of combining MAVE results with low‐throughput experiments to facilitate MAVE score interpretation and supervised model training.

Subjects

MeSH terms (MEDLINE)

ProteinsSupervised Machine LearningMutagenesis

Author keywords

deep mutational scanningmachine learningprotein abundanceprotein stabilityvariant effects

Substances

Proteins

Authors and institutions

  • The Linderstrøm-Lang Centre for Protein Science, Department of Biology, University of Copenhagen, Copenhagen, Denmark.

ORCID profiles: Kresten Lindorff-Larsen

Funding

  • Novo Nordisk Fonden

Cite

APA

Schulze, T. K., Blaabjerg, L. M., Cagiada, M., & Lindorff-Larsen, K. (2026). Supervised learning of protein variant effects across large‐scale mutagenesis datasets. Protein Science, 35(4). https://doi.org/10.1002/pro.70526

Vancouver

Schulze TK, Blaabjerg LM, Cagiada M, Lindorff-Larsen K. Supervised learning of protein variant effects across large‐scale mutagenesis datasets. Protein Science. 2026;35(4). doi:10.1002/pro.70526

BibTeX
@article{Schulze2026,
  author = {Schulze, Thea K and Blaabjerg, Lasse M and Cagiada, Matteo and Lindorff-Larsen, Kresten},
  title = {Supervised learning of protein variant effects across large‐scale mutagenesis datasets},
  journal = {Protein Science},
  year = {2026},
  volume = {35},
  number = {4},
  doi = {10.1002/pro.70526}
}
Start for free →