{
  "abstract": "Routine histology preserves tissue architecture but leaves its molecular organization largely unmeasured. Recovering this information requires learning across heterogeneous gene panels and constraining a biologically ill-posed morphology-to-expression mapping. Here we present Alpha Spatial (AS), a language-conditioned framework that predicts cell-level gene expression from histology while jointly learning from different tissues and panels. Matched biological context improved prediction over omitted context, whereas incompatible context reduced accuracy. Multi-tissue learning further improved breast-tissue prediction, and AS outperformed the evaluated methods while preserving molecular-state organization. In breast cancer, AS recovered POSTN-associated stromal programmes and HER2-associated molecular and spatial differences. Exploratory patient-level analyses identified a CXCL14-associated spatial state whose abundance was associated with non-response to trastuzumab-containing treatment and larger residual invasive tumours. These findings establish a framework for reusing heterogeneous molecular observations to investigate tissue organization and treatment-associated microenvironments in archived histology.",
  "abstract_shown": true,
  "affiliations": [
    "Tsinghua University"
  ],
  "article_type": "New results",
  "arxiv": null,
  "author_total": 9,
  "authors": [
    {
      "affs": [
        0
      ],
      "name": "X. YANG"
    },
    {
      "name": "Z. GUO"
    },
    {
      "name": "X. GUO"
    },
    {
      "name": "W. WANG"
    },
    {
      "name": "X. LIU"
    },
    {
      "name": "C. Zhang"
    },
    {
      "name": "B. Bai"
    },
    {
      "name": "Y. He"
    },
    {
      "name": "Y. LIU"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/12erysa18bz80.bib",
  "cited_by": 0,
  "doi": "10.64898/2026.10.08.756444",
  "fulltext_url": "https://www.biorxiv.org/content/10.64898/2026.10.08.756444v1.full",
  "id": "12erysa18bz80",
  "issns": [],
  "landing_url": "https://www.biorxiv.org/content/10.64898/2026.10.08.756444v1",
  "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": "W7222264349",
  "pdf_url": "https://www.biorxiv.org/content/10.64898/2026.10.08.756444v1.full.pdf",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-08",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/12erysa18bz80.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:27:59Z",
      "native_id": "10.64898/2026.10.08.756444",
      "source": "biorxiv",
      "url": "https://www.biorxiv.org/content/10.64898/2026.10.08.756444v1"
    }
  ],
  "subjects": [
    {
      "field": "biology",
      "origin": "biorxiv:bioinformatics",
      "subfield": "bioinformatics"
    }
  ],
  "title": "Alpha Spatial: Language-guided prediction of single-cell spatial gene expression from histology",
  "title_text": "Alpha Spatial: Language-guided prediction of single-cell spatial gene expression from histology",
  "type": "preprint",
  "updated_at": "2026-10-30T12:27:59Z",
  "url": "https://research.draconitic.com/paper/12erysa18bz80",
  "venue": "bioRxiv",
  "versions": []
}