Article

Learning Fragmentation Physics or Exploiting Sequence Priors? Benchmarking Bias in Deep Learning Models for De Novo Peptide Sequencing

2026-06-29

Abstract excerpt

Deep learning models have advanced de novo peptide sequencing, but their predictions may reflect both physics-based spectral evidence and learned peptide-sequence priors. Systematically measuring such prior-associated behavior is important for benchmarking model robustness beyond conventional proteomics data. Here, we introduce the Prior Bias Index (PBI), a general framework for measuring the extent to which model...

Topics

Open a Topic to create a Post that cites this publication.

Identifiers and source

Literature Corpus work: 05198caf-5b0c-5ad9-a7ea-d525d25b7d29
DOI: 10.64898/2026.06.23.734131

Open publication

Related research

Semantic proximity does not establish scientific evidence.

Click a neighbor to travelStep 1 · 12 closest

Learning Fragmentation Physics or Exploiting Sequence Priors? Benchmarking Bias in Deep Learning Models for De Novo Peptide SequencingDOI 10.64898/2026.06.23.734131Read

De novo peptide sequencing with InstaNovo: Accurate, database-free peptide identification for large scale proteomics experimentsSemantic proximity Read

De novo peptide sequencing with InstaNovo: Diffusion-powered, peptide identification for large scale proteomics experimentsSemantic proximity Read

A High-Quality Acetylation Dataset Reveals Modest Data Requirements for Transfer Learning to Identify Little Studied Post-Translational ModificationsSemantic proximity Read

Protein Language Model Decoys for Target Decoy Competition in Proteomics: Quality Assessment and BenchmarksSemantic proximity Read

AbNovoBench: a resource and benchmarking platform for monoclonal antibody de novo sequencingSemantic proximity Read

diagFDR: Verifiable FDR Reporting in Proteomics via Scope, Calibration, and Stability DiagnosticsSemantic proximity Read

PeptiDIA: A Machine Learning Framework for Enhanced Peptide Identification in Fast-Gradient Data-Independent Acquisition ProteomicsSemantic proximity Read

LipiDetective a deep learning model for the identification of molecular lipid species in tandem mass spectraSemantic proximity Read

Model Agnostic Semi-Supervised Meta-Learning Elucidates Understudied Out-of-distribution Molecular InteractionsSemantic proximity Read

Neural Spectral Prediction for Structure Elucidation with Tandem Mass SpectrometrySemantic proximity Read

Pepid: a Highly Modifiable, Bioinformatics-Oriented Peptide Search EngineSemantic proximity Read

From variability to consensus: rescoring harmonizes peptide identification across diverse search engines and datasetsSemantic proximity Read