Pairwise Attention: Leveraging Mass Differences to Enhance De Novo Sequencing of Mass Spectra

peer-reviewed · Journal of Proteome Research · 2025

peer-reviewed · Journal of Proteome Research · 2025. Joel Lapin et al. A fundamental challenge in mass spectrometry-based proteomics is determining which peptide generated a given…
Date 2025-06-02
Type peer-reviewed
Venue Journal of Proteome Research
Publisher American Chemical Society (ACS)
Contribution algorithm
DOI 10.1021/acs.jproteome.5c00063
Citations (OpenAlex) 2
Venue 2-year citedness 3.48

Abstract

A fundamental challenge in mass spectrometry-based proteomics is determining which peptide generated a given MS2 spectrum. Peptide sequencing typically relies on matching spectra against a known sequence database, which in some applications is not available. Deep learning-based de novo sequencing can address this limitation by directly predicting peptide sequences from MS2 data. We have seen the application of the transformer architecture to de novo sequencing produce state-of-the-art results on the so-called nine-species benchmark. In this study, we propose an improved transformer encoder inspired by the heuristics used in the manual interpretation of spectra. We modify the attention mechanism with a learned bias based on pairwise mass differences, termed Pairwise Attention (PA). Adding PA improves average peptide precision at 100% coverage by 12.7% (5.9 percentage points) over our base transformer on the original nine-species benchmark. We have also achieved a 7.4% increase over the previously published model Casanovo. Our MS2 encoding strategy is largely orthogonal to other transformer-based models encoding MS2 spectra, enabling straightforward integration into existing deep-learning approaches. Our results show that integrating domain-specific knowledge into transformers boosts de novo sequencing performance.

Authors

  1. Joel Lapin · Technical University of Munich
  2. Alfred Nilsson · KTH Royal Institute of Technology
  3. Mathias Wilhelm · Technical University of Munich
  4. Lukas Käll · KTH Royal Institute of Technology

Methods and tools

Cites (12)

Cited by (1)

Seen in the charts

Back to the full map

Back to top