{
  "abstract": "In mass-spectrometry-based proteomics, a substantial proportion of acquired spectra remain unidentified, concealing a vast reservoir of biological variations commonly termed the \"dark proteome\", potentially obscuring sequence mutations, non-canonical peptides and post-translational modifications. Existing deep-learning methods largely rely on supervised training with peptide-spectrum matches (PSMs), limiting their ability to learn from the much larger pool of unannotated spectra. Here we introduce {pi}-SPECFormer, a mass-spectrometry foundation model pre-trained on 291 million spectra through a two-stage framework that combines self-supervised and PSMs-guided representation learning. {pi}-SPECFormer transfers effectively to de novo peptide sequencing, spectrum clustering and modification detection, consistently outperforming task-specific baselines. When applied to proteomic profiles from 103 individuals with lung adenocarcinoma, the model identified nine high-confidence mutated peptides, including six that were absent from the reference databases. {pi}-SPECFormer thus enables unbiased decoding of molecular complexity beyond the boundaries of the canonical proteome.",
  "abstract_shown": true,
  "affiliations": [
    "National Center for Protein Sciences (Beijing)"
  ],
  "article_type": "New results",
  "arxiv": null,
  "author_total": 18,
  "authors": [
    {
      "name": "T. Ling"
    },
    {
      "name": "T. Liu"
    },
    {
      "name": "F. Xu"
    },
    {
      "name": "Y. Zhang"
    },
    {
      "name": "J. Xia"
    },
    {
      "name": "X. Huang"
    },
    {
      "name": "S. Chen"
    },
    {
      "name": "L. Xie"
    },
    {
      "name": "R. Xu"
    },
    {
      "name": "T. Yang"
    },
    {
      "name": "K. Li"
    },
    {
      "name": "G. Wang"
    },
    {
      "name": "Z. Liang"
    },
    {
      "name": "J. Zhu"
    },
    {
      "name": "S. Z. Li"
    },
    {
      "name": "F. He"
    },
    {
      "name": "Y. Wang"
    },
    {
      "affs": [
        0
      ],
      "name": "C. Chang"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/b000fgmn9q6ye.bib",
  "cited_by": 0,
  "doi": "10.64898/2026.11.06.764812",
  "fulltext_url": "https://www.biorxiv.org/content/10.64898/2026.11.06.764812v1.full",
  "id": "b000fgmn9q6ye",
  "issns": [],
  "landing_url": "https://www.biorxiv.org/content/10.64898/2026.11.06.764812v1",
  "language": null,
  "licence": "CC-BY-NC-ND",
  "licence_label": "CC BY-NC-ND",
  "licence_source": "article",
  "licence_url": "https://creativecommons.org/licenses/by-nc-nd/4.0/",
  "metadata_licence": null,
  "native_subjects": [
    [
      "biorxiv",
      "bioinformatics"
    ]
  ],
  "openalex": null,
  "pdf_url": "https://www.biorxiv.org/content/10.64898/2026.11.06.764812v1.full.pdf",
  "pmcid": null,
  "pmid": null,
  "published": "2026-11-06",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/b000fgmn9q6ye.ris",
  "sources": [
    {
      "first_seen": "2026-11-10T00:23:31Z",
      "native_id": "10.64898/2026.11.06.764812",
      "source": "biorxiv",
      "url": "https://www.biorxiv.org/content/10.64898/2026.11.06.764812v1"
    }
  ],
  "subjects": [
    {
      "field": "biology",
      "origin": "biorxiv:bioinformatics",
      "subfield": "bioinformatics"
    }
  ],
  "title": "π-SPECFormer: A Mass Spectrometry-based Foundation Model for Decoding the Dark Proteome",
  "title_text": "π-SPECFormer: A Mass Spectrometry-based Foundation Model for Decoding the Dark Proteome",
  "type": "preprint",
  "updated_at": "2026-11-10T00:23:31Z",
  "url": "https://research.draconitic.com/paper/b000fgmn9q6ye",
  "venue": "bioRxiv",
  "versions": []
}