pUniFind: a unified large pre-trained deep learning model pushing the limit of mass spectra interpretation

preprint · arXiv · 2025

preprint · arXiv · 2025. Jiale Zhao et al. Deep learning has advanced mass spectrometry data interpretation, yet most models remain feature extractors…
Date 2025-06-30
Type preprint
Venue arXiv
Publisher arXiv
Contribution algorithm
DOI 10.48550/arXiv.2507.00087
Citations (OpenAlex) 0

Abstract

Deep learning has advanced mass spectrometry data interpretation, yet most models remain feature extractors rather than unified scoring frameworks. We present pUniFind, the first large-scale multimodal pre-trained model in proteomics that integrates end-to-end peptide-spectrum scoring with open, zero-shot de novo sequencing. Trained on over 100 million open search-derived spectra, pUniFind aligns spectral and peptide modalities via cross modality prediction and outperforms traditional engines across diverse datasets, particularly achieving a 42.6 percent increase in the number of identified peptides in immunopeptidomics. Supporting over 1,300 modifications, pUniFind identifies 60 percent more PSMs than existing de novo methods despite a 300-fold larger search space. A deep learning based quality control module further recovers 38.5 percent additional peptides including 1,891 mapped to the genome but absent from reference proteomes while preserving full fragment ion coverage. These results establish a unified, scalable deep learning framework for proteomic analysis, offering improved sensitivity, modification coverage, and interpretability.

Authors

  1. Jiale Zhao · Chinese Academy of Sciences, University of Chinese Academy of Sciences
  2. Pengzhi Mao · Chinese Academy of Sciences, University of Chinese Academy of Sciences
  3. Kaifei Wang · Chinese Academy of Sciences, University of Chinese Academy of Sciences
  4. Yiming Li · University of Chinese Academy of Sciences
  5. Yaping Peng · Chinese Academy of Sciences, University of Chinese Academy of Sciences
  6. Ranfei Chen · Chinese Academy of Sciences, University of Chinese Academy of Sciences
  7. Shuqi Lu · DP Technology Co., Ltd.
  8. Xiaohong Ji · DP Technology Co., Ltd.
  9. Jiaxiang Ding · Chinese Academy of Sciences, University of Chinese Academy of Sciences
  10. Xin Zhang · University of Chinese Academy of Sciences
  11. Yucheng Liao · Peking University
  12. Weinan E · AI for Science Institute, Peking University
  13. Weijie Zhang · DP Technology Co., Ltd.
  14. Han Wen · AI for Science Institute, DP Technology Co., Ltd., State Key Laboratory of Medical Proteomics
  15. Hao Chi · Chinese Academy of Sciences, University of Chinese Academy of Sciences

Methods and tools

  • pUniFind: Multimodal pre-trained transformer for mass spectra that unifies peptide-spectrum scoring and zero-shot de novo sequencing in a single model. Trained on >100M open-search-derived spectra; reports +60% PSMs over prior de novo methods with 1,300+ modifications supported, and a DL-based QC step that recovers 38.5% additional peptides.

Cites (9)

Seen in the charts

Back to the full map

Back to top