EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

AFToolkit: a framework for molecular modeling of proteins with AlphaFold-derived representations

Maria Sindeeva iD, Alexander Telepov iD, Nikita Ivanisenko iD, Tatiana Shashkova iD, Kuzma Khrabrov iD, Artem Tsypin iD, Artur Kadurin iD, Olga Kardymon iD

Briefings in Bioinformatics · 2025

DOI: 10.1093/bib/bbaf324 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Briefings in Bioinformatics, 2025. Verfasst von Maria Sindeeva und 7 weiteren Autor:innen. Beteiligte Einrichtung: Bioinformatics Group, AIRI, Moscow 121170, Russia. Hauptthemen laut MEDLINE: Proteins, Models, Molecular, Software. Die Arbeit nennt 63 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

A key challenge in protein engineering is understanding how mutations affect protein fitness and stability. Most of current state-of-the-art models fine-tune protein structure prediction or protein language models or even pretrain their own. Despite its widespread use within computational workflows, AlphaFold2 exhibits limited sensitivity in assessing the effects of amino acid point mutations on protein structure, thereby constraining its utility in sequence design and protein engineering. In this work, we propose a simple modification of AlphaFold2 inference that improves the model’s capacity to capture the structural impacts of amino acid mutations. We achieve this by discarding the multiple sequence alignment and masking the template in recycling stages. Moreover, we introduce AFToolkit, a framework that leverages the embeddings of the modified AlphaFold2 model and simple adapter models to solve multiple protein engineering tasks. In contrast to other methods, our approach does not require fine-tuning the AlphaFold2 model or pretraining a new model from scratch on large datasets. It also supports handling multiple mutations, insertions, and deletions by directly modifying the input protein sequence. The proposed approach achieves strong performance across established benchmarks in terms of Spearman correlation: $0.68$ on PTMul, $0.60$ on cDNA-indel, and $0.57$ on C380.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

ProteinsModels, MolecularSoftwareProtein FoldingProtein ConformationComputational BiologyMutationAlgorithmsProtein Engineering

Stichwörter der Autor:innen

protein engineeringprotein stabilityprotein–protein binding affinity

Substanzen

Proteins

Autor:innen und Institutionen

  • Bioinformatics Group, AIRI, Moscow 121170, Russia.

ORCID-Profile: Maria Sindeeva, Alexander Telepov, Nikita Ivanisenko, Tatiana Shashkova, Kuzma Khrabrov, Artem Tsypin, Artur Kadurin, Olga Kardymon

Zitieren

APA

Sindeeva, M., Telepov, A., Ivanisenko, N., Shashkova, T., Khrabrov, K., Tsypin, A., Kadurin, A., & Kardymon, O. (2025). AFToolkit: a framework for molecular modeling of proteins with AlphaFold-derived representations. Briefings in Bioinformatics. https://doi.org/10.1093/bib/bbaf324

Vancouver

Sindeeva M, Telepov A, Ivanisenko N, Shashkova T, Khrabrov K, Tsypin A, et al. AFToolkit: a framework for molecular modeling of proteins with AlphaFold-derived representations. Briefings in Bioinformatics. 2025. doi:10.1093/bib/bbaf324

BibTeX
@article{Sindeeva2025,
  author = {Sindeeva, Maria and Telepov, Alexander and Ivanisenko, Nikita and Shashkova, Tatiana and Khrabrov, Kuzma and Tsypin, Artem and Kadurin, Artur and Kardymon, Olga},
  title = {AFToolkit: a framework for molecular modeling of proteins with AlphaFold-derived representations},
  journal = {Briefings in Bioinformatics},
  year = {2025},
  doi = {10.1093/bib/bbaf324}
}
Infokarte | EUVIMED

Artikeldaten derzeit nicht verfügbar.

Recherche