定义癌症的空间生态型

• Research Highlight
• Published: 11 June 2026

Bioinformatics

• Lin Tang1

_Nature Methods_volume 23,page 1074 (2026) Cite this article

• 171 Accesses

• 4 Altmetric

• Metrics details

Cancer cells are not lone wolves. Fueled and influenced by complicated interactions with their peers and other cell types in the surrounding microenvironment, they progress, metastasize and respond to therapies. Such complexities naturally mandate taking a multicellular and multimodal approach. However, how to best take advantage of the many types of omics data available is far from clear. Aaron Newman at Stanford University, Aadel Chaudhuri at Mayo Clinic and their colleagues have now developed new machine learning methods to dissect the tumor microenvironment (TME).

The team used a large collection of spatial transcriptomics data they compiled from different cancer types to develop a method called Spatial EcoTyper. Specifically, they adapted a similarity network fusion approach to identify shared multicellular patterns in a common embedding across different samples. Non-negative matrix factorization further revealed nine robust spatial clusters that the team defined as ‘spatial ecotypes’ (SEs). They validated these SEs in independent datasets generated by different technical platforms and demonstrated their associations with specific spatial patterns and molecular and clinical features. As Newman notes, among many of their findings, “We were surprised by the existence of SE-specific expression signatures that were largely agnostic to cell type (termed consensus programs) and that distinguished SEs by acute stress response, wound healing, immunosuppression, neovascularization, interferon signaling and other biological hallmarks of the TME.”

This is a preview of subscription content, access via your institution

Access options

Access Nature and 54 other Nature Portfolio journals

Get Nature+, our best-value online-access subscription

$32.99 /30 days

cancel any time

Subscribe to this journal

Receive 12 print issues and online access

$259.00 per year

only $21.58 per issue

Buy this article

• Purchase on SpringerLink
• Instant access to the full article PDF.

USD 39.95

Prices may be subject to local taxes which are calculated during checkout

Additional access options:

• Log in
• Learn about institutional subscriptions
• Read our FAQs
• Contact customer support

Author information

Authors and Affiliations

1. Nature Methods https://www.nature.com/nmeth/

Lin Tang

Authors

1. Lin Tang

Corresponding author

Correspondence to Lin Tang.

About this article

Cite this article

Tang, L. Defining cancer spatial ecotypes. _Nat Methods_23, 1074 (2026). https://doi.org/10.1038/s41592-026-03135-5

Download citation

• Published: 11 June 2026

• Version of record: 11 June 2026

• Issue date: June 2026

• DOI: https://doi.org/10.1038/s41592-026-03135-5