EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Artificial intelligence and first‐principle methods in protein redesign: A marriage of convenience?

Damiano Cianferoni iD, David Vizarraga iD, Ana María Fernández-Escamilla, Ignacio Fita, Rahma Hamdani iD, Raul Reche, Javier Delgado, Luis Serrano

Protein Science · 2025

DOI: 10.1002/pro.70210 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Protein Science, 2025. Verfasst von Damiano Cianferoni und 7 weiteren Autor:innen. Beteiligte Einrichtung: Centre for Genomic Regulation (CRG), The Barcelona Institute for Science and Technology, Barcelona, Spain. Hauptthemen laut MEDLINE: Proteins, Artificial Intelligence, Protein Engineering. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 56 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: Generalitat Valenciana, "la Caixa" Foundation, MCIN/ AEI / FEDER, UE.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Since AlphaFold2's rise, many deep learning methods for protein design have emerged. Here, we validate widely used and recognized tools, compare them with first‐principle methods, and explore their combinations, focusing on their effectiveness in protein redesign and potential for therapeutic repurposing. We address two challenges: evaluating tools and combinations ability to detect the effects of multiple concurrent mutations in protein variants, and leveraging large‐scale datasets to compare modeling‐free methods, namely force fields, which handle point mutations well with limited backbone rearrangement, and inverse folding tools, which excel at native sequence recovery but may struggle with non‐natural proteins. Debuting TriCombine, a tool that identifies residue triangles in input structures, matches them to a structural database, and scores mutants based on substitution frequencies, we shortlisted candidates, modeled them with FoldX, and generated 16 SH3 mutants carrying up to 9 concurrent substitutions. The dataset was expanded to include 36 mutants and 11 crystal structures (7 newly solved), along with a parallel set of multiple non‐concurrent mutants from three additional proteins. For broader validation, we analyzed 160,000 four‐site GB1 mutants and 163,555 (single and double) variants across 179 natural and de novo domains. We show that combining AI‐based modeling tools with force field scoring functions yields the most reliable results. Inverse folding tools perform very well but lose accuracy on less‐represented proteins. First‐principle force fields like FoldX remain the most accurate for point mutations. All methods perform worse when applied to unsolved de novo models, underscoring the need for hybrid strategies in robust protein design.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

ProteinsArtificial IntelligenceProtein EngineeringProtein FoldingModels, MolecularDeep LearningDatabases, ProteinProtein Conformation

Stichwörter der Autor:innen

artificial intelligencecrystallographic structureforce fieldprotein design

Substanzen

Proteins

Autor:innen und Institutionen

  • Centre for Genomic Regulation (CRG), The Barcelona Institute for Science and Technology, Barcelona, Spain.
  • Universitat de Barcelona (UB), Barcelona, Spain.
  • Instituto de Investigación, Desarrollo e Innovación en Biotecnología Sanitaria de Elche (IDiBE), Universitas Miguel Hernández, Elche, Spain.
  • Instituto de Biología Molecular de Barcelona (IBMB-CSIC), Parc Científic de Barcelona, Barcelona, Spain.
  • Universitat Pompeu Fabra (UPF), Barcelona, Spain.
  • ICREA, Barcelona, Spain.

ORCID-Profile: Damiano Cianferoni, David Vizarraga, Rahma Hamdani

Förderung

  • Generalitat Valenciana
  • "la Caixa" Foundation
  • MCIN/ AEI / FEDER, UE
  • Spanish National Plan for Scientific and Technical Research and Innovation
  • European Regional Development Fund
  • Centres de Recerca de Catalunya
  • Fundación Carmen y Severo Ochoa

Zitieren

APA

Cianferoni, D., Vizarraga, D., Fernández-Escamilla, A. M., Fita, I., Hamdani, R., Reche, R., Delgado, J., & Serrano, L. (2025). Artificial intelligence and first‐principle methods in protein redesign: A marriage of convenience?. Protein Science. https://doi.org/10.1002/pro.70210

Vancouver

Cianferoni D, Vizarraga D, Fernández-Escamilla AM, Fita I, Hamdani R, Reche R, et al. Artificial intelligence and first‐principle methods in protein redesign: A marriage of convenience?. Protein Science. 2025. doi:10.1002/pro.70210

BibTeX
@article{Cianferoni2025,
  author = {Cianferoni, Damiano and Vizarraga, David and Fernández-Escamilla, Ana María and Fita, Ignacio and Hamdani, Rahma and Reche, Raul and Delgado, Javier and Serrano, Luis},
  title = {Artificial intelligence and first‐principle methods in protein redesign: A marriage of convenience?},
  journal = {Protein Science},
  year = {2025},
  doi = {10.1002/pro.70210}
}
Kostenlos starten →