{
  "abstract": "Most molecular features detected by untargeted mass spectrometry-based metabolomics remain unannotated. This so-called 'dark matter' currently comprises more than 90% and likely holds uncharacterized bioactive chemistry. Structure prediction from fragmentation spectra remains an outstanding challenge, because a spectrum does not uniquely determine a structure and the space of candidate structures is mostly unbounded; correspondingly, existing methods rarely return the correct de novo structure as their top prediction. Here we describe De novo Ego network-Enabled Propagation and Neighborhood Extrapolation of Topology using Large Language Models (DEEPNET-LLM) reasoning approach, which uses Molecular Community Networking (MCN) to restrict the search of a query spectrum to its local neighborhood within the global metabolome, and applies AI reasoning to propagate structural information from neighbors to the query. For a collection of 900 authentic standards (2,249 spectra) that have never been publicly released to exclude possibility of data leakage, DEEPNET-LLM returned the correct structure as its top prediction for 84.7% of spectra. Accuracy was independent of precursor mass across a range over which the number of candidate structures grows by 30 orders of magnitude, and was comparable across ion polarity, adducts and in-source fragmentation. On a random subset of the public MassSpecGym benchmark, DEEPNET-LLM reached 85.9% top-1 accuracy excluding stereochemistry. We showcase the use of DEEPNET-LLM with an example of a discovery of tryptamine amidate of a bile acid in mouse gut, confirmed against a synthetic standard. The DEEPNET-LLM approach is orthogonal to existing in silico methods and can be combined with them.",
  "abstract_shown": true,
  "affiliations": [
    "University of Connecticut"
  ],
  "article_type": "New results",
  "arxiv": null,
  "author_total": 14,
  "authors": [
    {
      "name": "A. V. Melnik"
    },
    {
      "name": "A. M. P. Roberts"
    },
    {
      "name": "A. Semenov"
    },
    {
      "name": "D. Moradi"
    },
    {
      "name": "E. A. Coler"
    },
    {
      "name": "J. Goldstein"
    },
    {
      "name": "B. Ojha"
    },
    {
      "name": "N. J. Swift"
    },
    {
      "name": "S. Demirovic"
    },
    {
      "name": "M. W. Peczuh"
    },
    {
      "name": "T. D. Horvath"
    },
    {
      "name": "M. A. Engevik"
    },
    {
      "name": "V. Boginski"
    },
    {
      "affs": [
        0
      ],
      "name": "A. A. Aksenov"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/94yanqcmwq15e.bib",
  "cited_by": 0,
  "doi": "10.64898/2026.10.28.762558",
  "fulltext_url": "https://www.biorxiv.org/content/10.64898/2026.10.28.762558v1.full",
  "id": "94yanqcmwq15e",
  "issns": [],
  "landing_url": "https://www.biorxiv.org/content/10.64898/2026.10.28.762558v1",
  "language": null,
  "licence": "CC-BY-ND",
  "licence_label": "CC BY-ND",
  "licence_source": "article",
  "licence_url": "https://creativecommons.org/licenses/by-nd/4.0/",
  "metadata_licence": null,
  "native_subjects": [
    [
      "biorxiv",
      "bioinformatics"
    ]
  ],
  "openalex": null,
  "pdf_url": "https://www.biorxiv.org/content/10.64898/2026.10.28.762558v1.full.pdf",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-28",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/94yanqcmwq15e.ris",
  "sources": [
    {
      "first_seen": "2026-11-04T00:21:34Z",
      "native_id": "10.64898/2026.10.28.762558",
      "source": "biorxiv",
      "url": "https://www.biorxiv.org/content/10.64898/2026.10.28.762558v1"
    }
  ],
  "subjects": [
    {
      "field": "biology",
      "origin": "biorxiv:bioinformatics",
      "subfield": "bioinformatics"
    }
  ],
  "title": "DEEPNET-LLM: Highly accurate structure prediction of unknown metabolites from mass spectra via their molecular neighborhoods",
  "title_text": "DEEPNET-LLM: Highly accurate structure prediction of unknown metabolites from mass spectra via their molecular neighborhoods",
  "type": "preprint",
  "updated_at": "2026-11-05T00:21:01Z",
  "url": "https://research.draconitic.com/paper/94yanqcmwq15e",
  "venue": "bioRxiv",
  "versions": []
}