Abstract
Spatial transcriptomics (ST) technologies provide genome-wide transcriptomic profiles in tissue context but lack direct protein-level measurements, which are critical for interpreting cellular function and microenvironmental organization. To bridge this gap, we develop DGAT (Dual-Graph Attention Network), a deep learning framework that imputes spatial protein expression from ST data by learning RNA–protein relationships from spatial transcriptomic and proteomic datasets. The model constructs heterogeneous graphs integrating transcriptomic, proteomic, and spatial information, encoded using graph attention networks. Task-specific decoders reconstruct mRNA and predict protein abundance from a shared latent representation. Benchmarking across public and in-house datasets demonstrates that DGAT outperforms existing methods in protein imputation accuracy. Applied to ST datasets lacking protein measurements, the framework reveals spatially distinct cell states, immune phenotypes, and tissue architectures not evident from transcriptomics alone. Here, we show that this framework accurately reconstructs spatial protein landscapes, reveals biologically meaningful tissue organization, and enables protein-level interpretation from transcriptomics-only spatial data.
Similar content being viewed by others
Acknowledgements
We thank Xiaojun Ma and Annalisa Barrata for helpful feedback on the manuscript. Histology sectioning was performed by the Pitt Biospecimen Core, and 10x Visium library preparation and Illumina sequencing were conducted by the Health Sciences Sequencing Core at UPMC Children’s Hospital of Pittsburgh. Additional support was provided by the University of Pittsburgh, the Office of the Senior Vice Chancellor for Health Sciences, the Department of Pediatrics, and the Richard King Mellon Foundation for Pediatric Research.
Funding
This work was supported by the National Institutes of Health (R35GM146989 and R21CA294196) and by the Centers for Disease Control and Prevention, in association with the National Institute for Occupational Safety and Health and the National Mesothelioma Virtual Bank (U24OH009077). Computational analyses were supported by the University of Pittsburgh Center for Research Computing and the Extreme Science and Engineering Discovery Environment through the Bridges-2 system at the Pittsburgh Supercomputing Center.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Wang, H., Cody, B., Saavedra, M. et al. DGAT: a dual-graph attention network for inferring spatial protein landscapes from transcriptomics. Nat Commun (2026). https://doi.org/10.1038/s41467-026-73114-z
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41467-026-73114-z
