EUVIMEDEUVIMED entdecken
Open Access

How well do contextual protein encodings learn structure, function, and evolutionary context?

Sai Pooja Mahajan, Fátima A Dávila-Hernández, Jeffrey A Ruffolo, Jeffrey J Gray

Cell Systems · 2025

DOI: 10.1016/j.cels.2025.101201 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Cell Systems, 2025. Verfasst von Sai Pooja Mahajan und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Chemical and Biomolecular Engineering, Johns Hopkins University, Baltimore, MD 21218, USA. Electronic address: saipooja@gmail.com. Hauptthemen laut MEDLINE: Proteins. Die Arbeit nennt 62 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NIGMS NIH HHS.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In proteins, the optimal residue at any position is determined by its structural, evolutionary, and functional contexts-much like how a word may be inferred from its context in language. We trained masked label prediction models to learn representations of amino acid residues in different contexts. We focus questions on evolution and structural flexibility and whether and how contextual encodings derived through pretraining and fine-tuning may improve representations for specialized contexts. Sequences sampled from our learned representations fold into template structure and reflect sequence variations seen in related proteins. For flexible proteins, sampled sequences traverse the full conformational space of the native sequence, suggesting that plasticity is encoded in the template structure. For protein-protein interfaces, generated sequences replicate wild-type binding energies across diverse interfaces and binding strengths in silico. For the antibody-antigen interface, fine-tuning recapitulate conserved sequence patterns, while pretraining on general contexts improves sequence recovery for the hypervariable H3 loop. A record of this paper's transparent peer review process is included in the supplemental information.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

ProteinsEvolution, MolecularProtein ConformationHumansAmino Acid SequenceProtein BindingModels, Molecular

Stichwörter der Autor:innen

antibody designbinder designcontext-aware designdeep learningequivariant graph transformersfinetuningmasked modelspretraining protein modelsprotein designprotein flexibility

Substanzen

Proteins

Autor:innen und Institutionen

  • Department of Chemical and Biomolecular Engineering, Johns Hopkins University, Baltimore, MD 21218, USA. Electronic address: saipooja@gmail.com.
  • Department of Chemical and Biomolecular Engineering, Johns Hopkins University, Baltimore, MD 21218, USA.
  • Program in Molecular Biophysics, Johns Hopkins University, Baltimore, MD 21218, USA.
  • Department of Chemical and Biomolecular Engineering, Johns Hopkins University, Baltimore, MD 21218, USA; Program in Molecular Biophysics, Johns Hopkins University, Baltimore, MD 21218, USA; Johns Hopkins Data Science an…

Förderung

  • NIGMS NIH HHS

Zitieren

APA

Mahajan, S. P., Dávila-Hernández, F. A., Ruffolo, J. A., & Gray, J. J. (2025). How well do contextual protein encodings learn structure, function, and evolutionary context?. Cell Systems. https://doi.org/10.1016/j.cels.2025.101201

Vancouver

Mahajan SP, Dávila-Hernández FA, Ruffolo JA, Gray JJ. How well do contextual protein encodings learn structure, function, and evolutionary context?. Cell Systems. 2025. doi:10.1016/j.cels.2025.101201

BibTeX
@article{Mahajan2025,
  author = {Mahajan, Sai Pooja and Dávila-Hernández, Fátima A and Ruffolo, Jeffrey A and Gray, Jeffrey J},
  title = {How well do contextual protein encodings learn structure, function, and evolutionary context?},
  journal = {Cell Systems},
  year = {2025},
  doi = {10.1016/j.cels.2025.101201}
}
Kostenlos starten →