{
  "abstract": "Spatial transcriptomics captures molecular states within cells and their organisation in tissue. However, integrating fine-grained gene information with spatial context at scale remains challenging for existing foundation models. Here we present NexuST, a hierarchical foundation model that repeatedly interleaves gene-level molecular modelling with cell-level spatial modelling, allowing the two levels to refine one another during end-to-end pretraining. For pretraining, we curated HumanST-46M, comprising 45.7 million human cells from 72 datasets across 11 organs and three imaging-based platforms. Across four held-out datasets totalling approximately 2.6 million cells, NexuST achieved state-of-the-art or competitive performance in cell-type annotation, region prediction, gene recovery and neighbourhood-composition prediction. We find that cell-intrinsic expression remains informative even for spatial tasks, as shown by an expression-only PCA baseline, while NexuST shows particularly strong gains where spatial context is essential. Overall, NexuST establishes a hierarchical framework that can serve as a general backbone for future spatial transcriptomics foundation models.",
  "abstract_shown": true,
  "affiliations": [
    "The University of Manchester"
  ],
  "article_type": "New results",
  "arxiv": null,
  "author_total": 12,
  "authors": [
    {
      "name": "H. Liu"
    },
    {
      "name": "Q. Zhao"
    },
    {
      "name": "L. Lin"
    },
    {
      "name": "Z. Zou"
    },
    {
      "name": "W. Cai"
    },
    {
      "name": "J. Sun"
    },
    {
      "affs": [
        0
      ],
      "name": "Y. Zhou"
    },
    {
      "name": "M. A. Alvarez"
    },
    {
      "name": "A. Gilmore"
    },
    {
      "name": "M. Rattray"
    },
    {
      "name": "A. F. Frangi"
    },
    {
      "affs": [
        0
      ],
      "name": "H. Zhou"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/ekh88ghckmgm9.bib",
  "cited_by": 0,
  "doi": "10.64898/2026.09.22.753590",
  "fulltext_url": "https://www.biorxiv.org/content/10.64898/2026.09.22.753590v1.full",
  "id": "ekh88ghckmgm9",
  "issns": [],
  "landing_url": "https://www.biorxiv.org/content/10.64898/2026.09.22.753590v1",
  "language": null,
  "licence": "CC-BY",
  "licence_label": "CC BY",
  "licence_source": "article",
  "licence_url": "https://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": null,
  "native_subjects": [
    [
      "biorxiv",
      "bioinformatics"
    ],
    [
      "openalex",
      "1312|Single-cell and spatial transcriptomics"
    ]
  ],
  "openalex": "W7214670859",
  "pdf_url": "https://www.biorxiv.org/content/10.64898/2026.09.22.753590v1.full.pdf",
  "pmcid": null,
  "pmid": null,
  "published": "2026-09-22",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/ekh88ghckmgm9.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:24:55Z",
      "native_id": "10.64898/2026.09.22.753590",
      "source": "biorxiv",
      "url": "https://www.biorxiv.org/content/10.64898/2026.09.22.753590v1"
    }
  ],
  "subjects": [
    {
      "field": "biology",
      "origin": "biorxiv:bioinformatics",
      "subfield": "bioinformatics"
    }
  ],
  "title": "NexuST: A Hierarchical Foundation Model for Spatial Transcriptomics",
  "title_text": "NexuST: A Hierarchical Foundation Model for Spatial Transcriptomics",
  "type": "preprint",
  "updated_at": "2026-10-30T12:24:55Z",
  "url": "https://research.draconitic.com/paper/ekh88ghckmgm9",
  "venue": "bioRxiv",
  "versions": []
}