EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

PowerNovo2: A generative flow-based approach to non-autoregressive de novo peptide sequencing

Denis V Petrovskiy, Kirill S Nikolsky iD, Vladimir R Rudnev, Liudmila I Kulikova iD, Tatiana V Butkova iD, Kristina A Malsagova, Arthur T Kopylov, Anna L Kaysheva iD

PLOS Computational Biology · 20.05.2026 · 22 (5) · e1014298

DOI: 10.1371/journal.pcbi.1014298 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS Computational Biology, online erschienen am 20.05.2026. Verfasst von Denis V Petrovskiy und 7 weiteren Autor:innen. Beteiligte Einrichtung: Institute of Biomedical Chemistry, Moscow, Russia. Hauptthemen laut MEDLINE: Peptides, Proteomics, Sequence Analysis, Protein, Software. Die Arbeit nennt 22 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Program for Basic Research in the Russian Federation for a long-term period.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Proteomics utilizes tandem mass spectrometry (MS/MS) to determine peptide sequences, traditionally through database searches constrained by prior knowledge. De novo sequencing offers a database-free alternative but struggles with accurately modeling complex MS/MS spectra. Most current tools use autoregressive decoding, which is prone to error propagation and computationally slow. Here we present PowerNovo2, a non-autoregressive model based on generative normalizing flows. By leveraging variational inference, it effectively captures intricate token dependencies and peptide-level uncertainties. PowerNovo2 outperforms existing de novo tools in accuracy and speed, matching state-of-the-art autoregressive models like Casanovo while being 4.3 times faster. It also demonstrates competitive performance against other non-autoregressive methods such as π-PrimeNovo, particularly on long peptides and low-resolution spectra. As the first flow-based de novo sequencer, PowerNovo2 provides a scalable, accurate solution for large-scale proteomic applications.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

PeptidesProteomicsSequence Analysis, ProteinSoftwareTandem Mass SpectrometryAlgorithmsComputational BiologyDatabases, Protein

Substanzen

Peptides

Autor:innen und Institutionen

  • Institute of Biomedical Chemistry, Moscow, Russia.

ORCID-Profile: Kirill S Nikolsky, Liudmila I Kulikova, Tatiana V Butkova, Anna L Kaysheva

Förderung

  • Program for Basic Research in the Russian Federation for a long-term period

Zitieren

APA

Petrovskiy, D. V., Nikolsky, K. S., Rudnev, V. R., Kulikova, L. I., Butkova, T. V., Malsagova, K. A., Kopylov, A. T., & Kaysheva, A. L. (2026). PowerNovo2: A generative flow-based approach to non-autoregressive de novo peptide sequencing. PLOS Computational Biology, 22(5), e1014298. https://doi.org/10.1371/journal.pcbi.1014298

Vancouver

Petrovskiy DV, Nikolsky KS, Rudnev VR, Kulikova LI, Butkova TV, Malsagova KA, et al. PowerNovo2: A generative flow-based approach to non-autoregressive de novo peptide sequencing. PLOS Computational Biology. 2026;22(5):e1014298. doi:10.1371/journal.pcbi.1014298

BibTeX
@article{Petrovskiy2026,
  author = {Petrovskiy, Denis V and Nikolsky, Kirill S and Rudnev, Vladimir R and Kulikova, Liudmila I and Butkova, Tatiana V and Malsagova, Kristina A and Kopylov, Arthur T and Kaysheva, Anna L},
  title = {PowerNovo2: A generative flow-based approach to non-autoregressive de novo peptide sequencing},
  journal = {PLOS Computational Biology},
  year = {2026},
  volume = {22},
  number = {5},
  pages = {e1014298},
  doi = {10.1371/journal.pcbi.1014298}
}
Kostenlos starten →