Spatially guided translation from histology images to transcriptomic profiles using foundation model-driven contrastive learning
PLOS Computational Biology · 2026
DOI: 10.1371/journal.pcbi.1014762 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS Computational Biology, 2026. Verfasst von Zi Huai Huang und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Biochemistry, Western University, London, Canada. Hauptthemen laut MEDLINE: Graph Neural Networks, Spatial Transcriptomics. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Canada Research Chairs, Natural Sciences and Engineering Research Council of Canada, The Canadian Foundation for Innovation (CFI).
Abstract (Originalsprache)
Spatial transcriptomics (ST) enhances single-cell RNA sequencing by revealing transcript distribution, offering critical insights into heterogeneous diseases such as breast cancer. However, the high cost and lengthy processes of generating high-quality ST data limit clinical application. Recent deep learning methods predict ST from histology images, but often fail to capture both morphological features and spatial context. We introduce FOCST, a foundation model-driven framework for ST imputation that leverages spatial guided contrastive learning. FOCST begins with UNI, a large histopathology foundation model, to extract visual features from tissue images. These are integrated with expression data in a unified embedding space via contrastive learning, enabling cross-modal prediction and imputation. To further enhance spatial awareness, a graph neural network incorporates positional information, improving regional detection and interpretability.Benchmarking demonstrates FOCST’s superior performance over state-of-the-art methods and alternative vision encoders (paired Wilcoxon signed-rank tests, FDR-adjusted p < 0.05, N = 6 images). Predicted profiles enable clinically relevant downstream analyses, including patient stratification by treatment response (ROC AUC (Receiver Operating Characteristic – Area Under the Curve) = 0.79). Our results highlight the promise of combining foundation models and spatially guided learning to efficiently generate ST insights, advancing cancer research and precision medicine.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Department of Biochemistry, Western University, London, Canada.
- Department of Computer Sciences, Western University, London, Canada.
- Division of Biostatistics, University of Toronto, Toronto, Canada.
ORCID-Profile: Pingzhao Hu
Förderung
- Canada Research Chairs
- Natural Sciences and Engineering Research Council of Canada
- The Canadian Foundation for Innovation (CFI)
- Breast Cancer Canada
- CIHR
- Canadian Foundation for Innovation
- Canadian Institutes of Health Research
Zitieren
APA
Huang, Z. H., Xu, Z., & Hu, P. (2026). Spatially guided translation from histology images to transcriptomic profiles using foundation model-driven contrastive learning. PLOS Computational Biology. https://doi.org/10.1371/journal.pcbi.1014762
Vancouver
Huang ZH, Xu Z, Hu P. Spatially guided translation from histology images to transcriptomic profiles using foundation model-driven contrastive learning. PLOS Computational Biology. 2026. doi:10.1371/journal.pcbi.1014762
BibTeX
@article{Huang2026,
author = {Huang, Zi Huai and Xu, Ziyang and Hu, Pingzhao},
title = {Spatially guided translation from histology images to transcriptomic profiles using foundation model-driven contrastive learning},
journal = {PLOS Computational Biology},
year = {2026},
doi = {10.1371/journal.pcbi.1014762}
}
EUVIMED