PepPr: De Novo peptide sequencing with peptide priors

preprint · bioRxiv · 2026

preprint · bioRxiv · 2026. Samuel Khan et al. Mass spectrometry-based peptide sequencing underpins proteomics. Database search confines to predefined…
Date 2026-10-28
Type preprint
Venue bioRxiv
Publisher Cold Spring Harbor Laboratory
Contribution algorithm
DOI 10.64898/2026.10.27.762126

Abstract

Mass spectrometry-based peptide sequencing underpins proteomics. Database search confines to predefined candidates, whereas de novo peptide sequencing (DNPS) struggles with incomplete spectra. Existing hybrid methods use databases only for post hoc correction or retrieve previously identified spectra to guide decoding. We instead combine DNPS with a peptide prior: a language model fitted to a reference proteome, tolerating variants yet uncertain on strongly divergent sequences. Relative to prior-free DNPS, our implementation, PepPr, improved reference peptide identification and improved rather than suppressed variant identification. Using a related proteome as prior, it matched or outperformed homology-driven database search, and its gain over prior-free DNPS grew with relatedness. In antibody sequencing, it improved recovery of peptides spanning the variable-constant boundary, deamidation calls, and assembly, while preserving recall of hypervariable regions inaccessible to database search. Peptide priors thus bring reference knowledge into DNPS without limiting discovery.

Authors

  1. Samuel Khan · Technical University of Munich
  2. Daniela Klaproth-Andrade · Technical University of Munich
  3. Yanik Bruns · Technical University of Munich
  4. Mathias Wilhelm · Technical University of Munich
  5. Julien Gagneur · Helmholtz Center Munich, Technical University of Munich

Methods and tools

  • PepPr: A peptide language model fitted to a reference proteome, fused through a small learned head with the next-residue logits of an autoregressive sequencer (Casanovo or ContraNovo) before beam search, so reference context fills gaps in incomplete spectra without excluding variants.

Methods it uses

Data deposited

Trained on

Evaluated on

Seen in the charts

Back to the full map

Back to top