NovoCRF: Globally Normalized De Novo Peptide Sequencing on a Prefix-Mass Lattice

preprint · arXiv · 2026

preprint · arXiv · 2026. Rui Qiao et al. De novo peptide sequencing has moved from dynamic programming over hand-designed or generative scores to…
Date 2026-11-03
Type preprint
Venue arXiv
Publisher arXiv
Contribution algorithm
DOI 10.48550/arXiv.2611.05260

Abstract

De novo peptide sequencing has moved from dynamic programming over hand-designed or generative scores to large transformers that learn the task end to end. We present NovoCRF, a conditional random field (CRF) whose states are prefix masses, so that its paths are exactly the peptides within +/-0.06 Da of the precursor mass. A small network scores each edge from physics-derived ion-match features and is trained with its scores globally normalized over all mass-feasible peptides. To our knowledge, NovoCRF is the first de novo method trained with exact normalization over all mass-feasible peptides. Our software has three modules, each with exactly one job: a precursor mass finder recovers the accurate precursor mass from MS1, the CRF finds the peptides that fit the precursor mass by exact k-best decoding, and two BiGRU rerankers add sequence context and internal-fragment evidence. The model has 310k parameters and trains in about 17 GPU-hours. On seven held-out data-dependent acquisition datasets published in 2026 (126,262 labeled spectra), NovoCRF’s mean peptide recall is 42.4%, against 39.1% for Casanovo v5 with five beams, and its amino-acid recall 64.0% against 53.1%, at 1,746 spectra/s on 18 CPU threads, 55x Casanovo’s throughput on a GPU. Split by fragment evidence, NovoCRF leads on labels with up to three peptide bonds without a b or y ion, and Casanovo only beyond. NovoCRF can also search modifications it was never trained on at their exact mass: it finds 32.1% of the hydroxyproline peptides of a lion bone sample.

Authors

  1. Rui Qiao · Bioinformatics Solutions Inc., University of Waterloo
  2. Lei Xin · Bioinformatics Solutions Inc., Western University
  3. Baozhen Shan · Bioinformatics Solutions Inc., Western University

Methods and tools

  • NovoCRF: A conditional random field whose states are prefix masses, so every path is a peptide within 0.06 Da of the precursor; a 310k-parameter scorer of physics-derived ion-match features is trained with exact global normalization over all mass-feasible peptides, decoded by exact k-best search and reranked by two BiGRUs.

Trained on

Evaluated on

Seen in the charts

Back to the full map

Back to top