# Tumour Histology and Multi-omics Explorer

> A curated catalogue of public tumour datasets that link histology with molecular measurements: TCGA patients with RNA-seq and DNA methylation (450k), and published spatial omics studies with matched H&E images. Release 0.11.1, 16 October 2026.

The selected TCGA analysis panel has 7,826 patients in 18 analysis cohorts, 6,561 with both RNA-seq and 450k methylation. 113 spatial omics resources are catalogued: 16 with the original H&E and matched molecular data located, and 29 more with matched molecular data and a preview image only. 22 cancer types have their own page. The web pages draw their tables with JavaScript; the Markdown and JSON files below hold the same content as text.

Read these points before using any count:

- The selected analysis panel is one selection for one study design. General availability per TCGA project is reported separately and is larger; do not mix the two.
- Spatial studies come from patients other than those in TCGA. They add tissue context, not measurements for TCGA patients.
- Spatial resources are shown at two image levels. "Original H&E located": the original scan, measured expression and coordinates were found. "Matched data with preview image": expression and coordinates match a released preview image, and the original scan was not located; such data support molecular spatial maps, not whole-slide or tile models, cell morphology, physical distances or registration sign-off. Registration, calibration and tissue QC may be pending at either level.
- Spatial section records are per study, not globally unique physical sections or independent assays. Patients are study-local IDs given by the source; sections without one add no patients, and no eligible-patient total is established.
- Counts reported by a study keep that study's own unit (sections, slides, samples, donors) and are never added across studies. An unknown count is absent, not zero.
- Normal tissue comes from patients with cancer, not healthy donors.
- There is no treated TCGA reference set: treatment-flagged patients were withheld, not collected as a treated group.
- The pooled glioma cohort (TCGA-LGG and TCGA-GBM) is a classification benchmark; integrated WHO CNS5 diagnoses are not adjudicated.
- Data remain at their original repositories. Listing a dataset does not mean access has been granted or requested.

Licence: the curated metadata and these text files are [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/); the code is GPL-3.0. The underlying datasets keep their own terms: cite the original publications and repositories for each dataset. Maintainer: Matthieu Foll (IARC); questions and corrections through [GitHub issues](https://github.com/IARCbioinfo/histology-multiomics-explorer/issues).

## Pages as Markdown

- [TCGA reference](https://iarcbioinfo.github.io/histology-multiomics-explorer/md/tcga.md): the selected analysis panel by cohort, the patient screen and coverage by TCGA project
- [Spatial studies](https://iarcbioinfo.github.io/histology-multiomics-explorer/md/spatial.md): spatial resources by cancer type and image level, the package and public-route reviews, and status, access and assay for each
- [Cancer types](https://iarcbioinfo.github.io/histology-multiomics-explorer/md/cancers/index.md): one page per cancer type with its TCGA data, spatial studies, selection history and gaps
- [Curation and methods](https://iarcbioinfo.github.io/histology-multiomics-explorer/methods.html): definitions of count units, linkage levels, exclusions, pooling and limitations (static HTML)
- [Research applications](https://iarcbioinfo.github.io/histology-multiomics-explorer/applications.html): study designs the data can support, with their limits (static HTML)

## Agent interfaces

- [MCP server](https://iarcbioinfo.github.io/histology-multiomics-explorer/agents.html#mcp): a read-only Model Context Protocol server with 10 tools (`catalogue_overview`, `get_cancer_type`, `search_resources`, `get_resource`, `search_spatial_sections`, `get_spatial_section`, `get_tcga_cohort`, `search_tcga_patients`, `get_tcga_patient`, `explain_term`). It runs on the user's computer: `uvx --from "git+https://github.com/IARCbioinfo/histology-multiomics-explorer#subdirectory=mcp" histology-multiomics-explorer-mcp` reads the published site, and `uv run --directory mcp histology-multiomics-explorer-mcp` reads a copy of the repository. Setup for Claude Code and Claude Desktop is on the linked page.
- [AGENTS.md](https://github.com/IARCbioinfo/histology-multiomics-explorer/blob/main/AGENTS.md): instructions for coding agents working on the [repository](https://github.com/IARCbioinfo/histology-multiomics-explorer)

## Data

- [Data index](https://iarcbioinfo.github.io/histology-multiomics-explorer/data/index.json): every data file with its description, record count, size, SHA-256 checksum and JSON Schema
- [Field reference](https://iarcbioinfo.github.io/histology-multiomics-explorer/data/README.md): meaning of every field and controlled vocabulary
- [Per-cancer JSON](https://iarcbioinfo.github.io/histology-multiomics-explorer/data/by-cancer/BRCA.json): one file per cancer page, `data/by-cancer/<CODE>.json`, with the cancer record, TCGA counts, spatial resources, located sections, other resources and evidence
- [Glossary](https://iarcbioinfo.github.io/histology-multiomics-explorer/data/glossary.json): Definitions of the terms used for counts and curation, and every display vocabulary of the site
- [release.json](https://iarcbioinfo.github.io/histology-multiomics-explorer/data/release.json): Release version, date and the dates on which sources were checked
- [cancers.json](https://iarcbioinfo.github.io/histology-multiomics-explorer/data/cancers.json): Cancer types: page codes, panel rationale, linked TCGA project and analysis cohort, histology strata, summary and gaps
- [evidence.json](https://iarcbioinfo.github.io/histology-multiomics-explorer/data/evidence.json): Published archetype studies used as the panel's selection history
- [cohorts.json](https://iarcbioinfo.github.io/histology-multiomics-explorer/data/cohorts.json): Studies and resources in every data layer, with image level, linkage, access, open checks, reported counts and the spatial, package and follow-up review details
- [captures.json](https://iarcbioinfo.github.io/histology-multiomics-explorer/data/captures.json): Located spatial transcriptomics tissue section records, one per study and section, with measured expression, coordinates and the original H&E or a preview image (image_level)
- [tcga_counts.json](https://iarcbioinfo.github.io/histology-multiomics-explorer/data/tcga_counts.json): TCGA molecular screen: patients with RNA-seq and 450k methylation per project, before and after assay-quality exclusions, with normal tissue
- [tcga_eligibility.json](https://iarcbioinfo.github.io/histology-multiomics-explorer/data/tcga_eligibility.json): TCGA selected analysis panel: counts per project after assay QC, after the patient screen and selected, and the 18 analysis cohorts
- [tcga_patients.json](https://iarcbioinfo.github.io/histology-multiomics-explorer/data/tcga_patients.json): Per-patient RNA-seq and 450k states (retained, excluded, unresolved, absent) for tumour and normal tissue
- [tcga_patient_review.json](https://iarcbioinfo.github.io/histology-multiomics-explorer/data/tcga_patient_review.json): Per-patient treatment and diagnosis screen, analysis cohort, study role and published annotations, as reason codes
- [tcga_flags.json](https://iarcbioinfo.github.io/histology-multiomics-explorer/data/tcga_flags.json): Every excluded or unresolved TCGA aliquot with its reason and the layers still retained
- [tcga_reviews.json](https://iarcbioinfo.github.io/histology-multiomics-explorer/data/tcga_reviews.json): Histology review of TCGA-ESCA and TCGA-PAAD cases, cohort definitions and the patient scope review
- [tcga_resources.json](https://iarcbioinfo.github.io/histology-multiomics-explorer/data/tcga_resources.json): Links to current GDC data, PanCanAtlas files, histology sources and reference publications
- [leads.json](https://iarcbioinfo.github.io/histology-multiomics-explorer/data/leads.json): Related leads that add no resources or patients: unassigned releases, precursor studies and reuse records

## Cancer pages

- [Breast invasive carcinoma](https://iarcbioinfo.github.io/histology-multiomics-explorer/md/cancers/BRCA.md) (`BRCA`): 1,062 TCGA patients in the selected panel; 8 spatial resources, 2 with original H&E located, 2 with matched data and a preview image
- [Colon adenocarcinoma](https://iarcbioinfo.github.io/histology-multiomics-explorer/md/cancers/COAD.md) (`COAD`): 427 TCGA patients in the selected panel, within pooled CRC; 11 spatial resources, 3 with original H&E located, 3 with matched data and a preview image
- [Rectal adenocarcinoma](https://iarcbioinfo.github.io/histology-multiomics-explorer/md/cancers/READ.md) (`READ`): 154 TCGA patients in the selected panel, within pooled CRC; 11 spatial resources, 3 with original H&E located, 3 with matched data and a preview image
- [Colorectal adenocarcinoma](https://iarcbioinfo.github.io/histology-multiomics-explorer/md/cancers/CRC.md) (`CRC`): 581 TCGA patients in the selected panel (pooled TCGA-COAD + TCGA-READ); 11 spatial resources, 3 with original H&E located, 3 with matched data and a preview image
- [Thyroid carcinoma](https://iarcbioinfo.github.io/histology-multiomics-explorer/md/cancers/THCA.md) (`THCA`): 495 TCGA patients in the selected panel; 9 spatial resources, 2 with original H&E located, 1 with matched data and a preview image
- [Bladder urothelial carcinoma](https://iarcbioinfo.github.io/histology-multiomics-explorer/md/cancers/BLCA.md) (`BLCA`): 388 TCGA patients in the selected panel; 10 spatial resources, 2 with original H&E located, 3 with matched data and a preview image
- [Lower-grade glioma](https://iarcbioinfo.github.io/histology-multiomics-explorer/md/cancers/LGG.md) (`LGG`): 502 TCGA patients in the selected panel, within pooled GLIOMA; 10 spatial resources, 1 with original H&E located, 4 with matched data and a preview image
- [Glioblastoma (historical TCGA diagnosis)](https://iarcbioinfo.github.io/histology-multiomics-explorer/md/cancers/GBM.md) (`GBM`): 225 TCGA patients in the selected panel, within pooled GLIOMA; 3 spatial resources, 1 with original H&E located, 1 with matched data and a preview image
- [Diffuse glioma classification benchmark](https://iarcbioinfo.github.io/histology-multiomics-explorer/md/cancers/GLIOMA.md) (`GLIOMA`): 727 TCGA patients in the selected panel (pooled TCGA-LGG + TCGA-GBM); 10 spatial resources, 1 with original H&E located, 4 with matched data and a preview image
- [Liver hepatocellular carcinoma](https://iarcbioinfo.github.io/histology-multiomics-explorer/md/cancers/LIHC.md) (`LIHC`): 357 TCGA patients in the selected panel; 17 spatial resources, 1 with original H&E located, 3 with matched data and a preview image
- [Lung adenocarcinoma](https://iarcbioinfo.github.io/histology-multiomics-explorer/md/cancers/LUAD.md) (`LUAD`): 457 TCGA patients in the selected panel; 11 spatial resources, 2 with original H&E located, 3 with matched data and a preview image
- [Head and neck squamous cell carcinoma](https://iarcbioinfo.github.io/histology-multiomics-explorer/md/cancers/HNSC.md) (`HNSC`): 514 TCGA patients in the selected panel; 13 spatial resources, 3 with original H&E located, 3 with matched data and a preview image
- [Cervical squamous cell carcinoma and endocervical adenocarcinoma](https://iarcbioinfo.github.io/histology-multiomics-explorer/md/cancers/CESC.md) (`CESC`): 249 TCGA patients in the selected panel, squamous cell carcinoma only (CSCC); 3 spatial resources, verification pending
- [Oesophageal carcinoma](https://iarcbioinfo.github.io/histology-multiomics-explorer/md/cancers/ESCA.md) (`ESCA`): 90 TCGA patients in the selected panel, squamous cell carcinoma only (ESCC); 3 spatial resources, verification pending
- [Kidney renal clear cell carcinoma](https://iarcbioinfo.github.io/histology-multiomics-explorer/md/cancers/KIRC.md) (`KIRC`): 494 TCGA patients in the selected panel; 4 spatial resources, 2 with matched data and a preview image
- [Kidney renal papillary cell carcinoma](https://iarcbioinfo.github.io/histology-multiomics-explorer/md/cancers/KIRP.md) (`KIRP`): 275 TCGA patients in the selected panel; 2 spatial resources, verification pending
- [Lung squamous cell carcinoma](https://iarcbioinfo.github.io/histology-multiomics-explorer/md/cancers/LUSC.md) (`LUSC`): 471 TCGA patients in the selected panel; 2 spatial resources, 1 with matched data and a preview image
- [Pheochromocytoma and paraganglioma](https://iarcbioinfo.github.io/histology-multiomics-explorer/md/cancers/PCPG.md) (`PCPG`): 178 TCGA patients in the selected panel; 2 spatial resources, verification pending
- [Prostate adenocarcinoma](https://iarcbioinfo.github.io/histology-multiomics-explorer/md/cancers/PRAD.md) (`PRAD`): 402 TCGA patients in the selected panel; 4 spatial resources, verification pending
- [Stomach adenocarcinoma](https://iarcbioinfo.github.io/histology-multiomics-explorer/md/cancers/STAD.md) (`STAD`): 414 TCGA patients in the selected panel; 2 spatial resources, 2 with matched data and a preview image
- [Uterine corpus endometrial carcinoma](https://iarcbioinfo.github.io/histology-multiomics-explorer/md/cancers/UCEC.md) (`UCEC`): 523 TCGA patients in the selected panel; 2 spatial resources, 2 with matched data and a preview image
- [Pancreatic ductal adenocarcinoma](https://iarcbioinfo.github.io/histology-multiomics-explorer/md/cancers/PDAC.md) (`PDAC`): 149 TCGA patients in the selected panel, ductal adenocarcinoma only (PDAC); 3 spatial resources, 2 with original H&E located, 1 with matched data and a preview image

## Optional

- [Sources](https://iarcbioinfo.github.io/histology-multiomics-explorer/sources.html): citations and access routes for every dataset
- [Data explorer](https://iarcbioinfo.github.io/histology-multiomics-explorer/explore.html): interactive filters and CSV or JSON export (needs JavaScript)