ROSIE: AI generation of multiplex immunofluorescence staining from histopathology images
Hematoxylin and eosin (H&E) is a common and inexpensive histopathology assay. Though widely used and information-rich, it cannot directly inform about specific molecular markers, which require additional experiments to assess. To address this gap, we present ROSIE, a deep-learning framework that computationally imputes the expression and localization of dozens of proteins from H&E images. Our model is trained on a dataset of over 1300 paired and aligned H&E and multiplex immunofluorescence (mIF) samples from over a dozen tissues and disease conditions, spanning over 16 million cells. Validation of our in silico mIF staining method on held-out H&E samples demonstrates that the predicted biomarkers are effective in identifying cell phenotypes, particularly distinguishing lymphocytes such as B cells and T cells, which are not readily discernible with H&E staining alone. Additionally, ROSIE facilitates the robust identification of stromal and epithelial microenvironments and immune cell subtypes like tumor-infiltrating lymphocytes (TILs), which are important for understanding tumor-immune interactions and can help inform treatment strategies in cancer research.
Authors
- James Zou (ORCID: https://orcid.org/0000-0001-8880-4764)
- Vivek Behera (ORCID: https://orcid.org/0000-0001-6013-5706)
- Nitya Thakkar (ORCID: https://orcid.org/0000-0002-5647-6775)
- Gregory W. Charville (ORCID: https://orcid.org/0000-0002-2774-2704)
- Christian M. Schürch (ORCID: https://orcid.org/0000-0002-1792-1768)
- Christopher I. Li (ORCID: https://orcid.org/0000-0003-1543-0743)
- Jeroen R. Huyghe (ORCID: https://orcid.org/0000-0001-6027-9806)
- Matthew Bieniosek
- Alexandro E. Trevino (ORCID: https://orcid.org/0000-0002-9763-1540)
- Aaron T. Mayer (ORCID: https://orcid.org/0000-0001-6429-0420)
- Ulrike Peters (ORCID: https://orcid.org/0000-0001-5666-9318)
- Zhenqin Wu (ORCID: https://orcid.org/0000-0002-0148-7055)
- Ahmad Makky (ORCID: https://orcid.org/0000-0002-4205-1303)
- Eric Q. Wu (ORCID: https://orcid.org/0000-0001-8379-3373)
- Hannah Giba
- Li Li (ORCID: https://orcid.org/0000-0002-8630-3699)
- Arjun S. Raman (ORCID: https://orcid.org/0000-0002-7303-7272)
Institutions
- SRI International (US)
- Ochsner Health System (US)
- University of Washington (US)
- University of Chicago (US)
- Fred Hutch Cancer Center (US)
- Menlo School (US)
- Enable Ireland (IE)
- Stanford Medicine (US)
- Enable Biosciences (United States) (US)
- University of Tübingen (DE)
- Stanford University (US)
Publication Details
- Journal
- Nature Communications
- Published
- 2025-08-16
- DOI
- https://doi.org/10.1038/s41467-025-62346-0
- Citations
- 11
- Primary Topic
- AI in cancer detection
- Type
- article
- Field-Weighted Citation Impact
- 27.29